Psychogenic purpura (Gardner-Diamond syndrome)

INTRODUCTION — Psychogenic purpura (also referred to as Gardner-Diamond syndrome, autoerythrocyte sensitization, or painful bruising syndrome) is a rare and poorly understood clinical presentation in which patients develop unexplained painful bruises, mostly on the extremities and/or face, during times of stress. The mechanism by which stress induces these lesions is unclear. It has been speculated that this might be the disease entity responsible for historical descriptions of religious stigmata.

The clinical manifestations, diagnosis, and management of psychogenic purpura are discussed here. Related discussions are presented in separate topic reviews:

●Approach to a suspected bleeding disorder – (See "Approach to the adult with a suspected bleeding disorder".)

●Diagnosis of von Willebrand disease – (See "Clinical presentation and diagnosis of von Willebrand disease".)

●Functional neurological symptom disorder (conversion disorder) – (See "Functional neurological symptom disorder (conversion disorder) in adults: Terminology, diagnosis, and differential diagnosis".)

●Somatic symptom and related disorders – (See "Somatic symptom disorder: Assessment and diagnosis".)

TERMINOLOGY AND GENERAL CONCEPTS — Psychogenic purpura describes one or more episodes of purpura (confluent areas of superficial, non-blanching skin bleeding) that are not due to an underlying bleeding disorder or trauma to the affected area. Images are included with the discussion below. (See 'Painful purpura' below.)

Psychogenic purpura is a diagnosis of exclusion. Before making the diagnosis, clinicians should rule out bleeding disorders; skin, soft tissue, and vascular conditions; trauma; and self-inflicted injury due to psychiatric conditions. The extent of the evaluation is individualized depending on the severity of bleeding and other patient characteristics, as described below. (See 'Evaluation' below.)

Clinicians may reasonably wonder when the diagnosis of psychogenic purpura is appropriate, given that it is extremely rare and unexplained. There is no simple answer, as the condition is poorly understood. However, there is a reasonably well-stereotyped syndrome described in the literature and we believe this is an appropriate diagnosis in rare, selected cases that match the patterns discussed herein and are not explained by another diagnosis. (See 'Diagnosis' below.)

Psychogenic purpura is not classified by name in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), and it does not fit any broad DSM-5 diagnosis. The following psychiatric disorders may help in conceptualizing the underpinnings of some cases of psychogenic purpura, although this is not meant to imply that patients with psychogenic purpura require a psychiatric diagnosis:

●"Conversion-like disorder" – Psychogenic purpura shares features with functional neurological symptom disorders (conversion disorders), which are classified in the DSM-5 as disorders with one or more neurologic symptoms that are incompatible with a medical or neurologic disease. Examples include psychogenic nonepileptic seizures or functional movement disorders.

Psychogenic purpura does not have neurologic features and thus does not meet the definition of functional neurological symptom disorder (conversion disorder). However, psychogenic purpura manifests in clinical signs that are presumed not to be intentionally produced by the patient, and its mechanism of action is unknown. Thus, psychogenic purpura can be understood as a "conversion-like disorder."

While stress is part of the definition of psychogenic purpura, proximate stressors are no longer seen as necessary to the manifestation or diagnosis of functional neurological symptom disorder. (See "Functional neurological symptom disorder (conversion disorder) in adults: Terminology, diagnosis, and differential diagnosis" and "Functional neurological symptom disorder (conversion disorder) in adults: Clinical features, assessment, and comorbidity".)

●Somatic symptom and illness anxiety disorders – Psychogenic purpura may sometimes have features that overlap with somatic symptom disorder (SSD). SSD is defined by maladaptive illness behaviors rather than by the absence of a discernable etiology. (See "Somatic symptom disorder: Assessment and diagnosis" and 'Level of distress and/or psychiatric comorbidities' below.)

The overlap between psychogenic purpura and somatic symptom and illness anxiety disorders may be observed at the differential diagnosis or postdiagnosis stage of care. In somatic symptom disorder, a provider might note and separately address distress (cosmetic concern, pain response) that is disproportionate to objective signs or far in excess of that seem of most "typical" patients. In illness anxiety disorder, the provider's attention would be drawn to a patient's excessive fears about underlying causes of the illness or need for caution to avoid injury.

In functional neurological symptom disorder, patients are no more or less distressed by their symptoms than patients with analogous organic diagnoses. In contrast, somatic symptom and illness anxiety disorders are defined by the nature and expression of patients' illness behaviors. Some individuals with psychogenic purpura are distressed by their symptoms and others are not (or appear not to be). (See 'Level of distress and/or psychiatric comorbidities' below.)

If there is strong suspicion or proof that the patient is intentionally producing symptoms or signs for immaterial or material gain, their presentation should be identified as factitious disorder or malingering, respectively, rather than psychogenic purpura. (See 'Differential diagnosis' below.)

It may be useful to think of psychogenic purpura analogously with other conditions in which physical changes are documented in individuals who have no underlying medical etiology for these findings.

●Pseudocyesis – Pseudocyesis (also called false pregnancy) refers to the appearance of signs and symptoms of pregnancy in an individual who is not pregnant; it is classified in DSM-5 under somatic symptom and related disorders. (See "Pseudocyesis".)

●Stress-induced (takotsubo) cardiomyopathy – Stress-induced cardiomyopathy (including takotsubo cardiomyopathy or broken heart syndrome) refers to transient global left ventricular dysfunction in the absence of significant coronary artery disease or other explanations for wall motion abnormalities. Like psychogenic purpura, takotsubo cardiomyopathy may be caused by emotional or physical stress. Unlike psychogenic purpura, takotsubo cardiomyopathy has a proposed mechanism involving catecholamine effects. Stress-induced cardiomyopathy is not classified in the DSM-5. (See "Clinical manifestations and diagnosis of stress (takotsubo) cardiomyopathy".)

POSSIBLE MECHANISMS — The pathophysiology of psychogenic purpura is not understood, and the rarity of the condition and lack of preclinical model systems make studies of its mechanisms difficult to conduct. It is even difficult to definitively state that psychogenic purpura is a homogenous category rather than the observed manifestation of multiple different pathophysiologies. Nonetheless, stress-induced hemostatic changes and immunologic effects may play a role [1,2].

Hemostatic changes — It is possible that psychologic stress alters the levels of certain proteins involved in the hemostatic process. The following observations apply generally to hemostasis and are not specific to psychogenic purpura:

●Stress-induced increases in endogenous glucocorticoids are known to occur [3], and therapeutic doses of glucocorticoids are known to cause thinning of the skin and increased skin bleeding. However, changes in circulating glucocorticoid levels have not been demonstrated in psychogenic purpura, and the timing of steroid-induced cutaneous atrophy in individuals on long-term glucocorticoid therapy does not match with the rapid appearance of purpura seen in psychogenic purpura. (See "Major adverse effects of systemic glucocorticoids", section on 'Dermatologic effects and appearance'.)

●Antifibrinolytic proteins such as tissue plasminogen activator (tPA) may contribute to bleeding. In one report, an increased intradermal tPA concentration could be shown in immunoassays of sites of psychogenic purpura [4]. There is speculation that tPA release from endothelial cells is triggered by neuropeptides in the skin [3,4]. However, the patients in some case reports that demonstrated fibrinolytic defects also described significant bleeding (eg, joint bleeding), calling into question that the diagnosis was actually psychogenic purpura [4,5].

●Levels of von Willebrand factor (VWF) have been observed to increase when an individual is under stress (physiologic or emotional). This change would be expected to reduce rather than increase bleeding. However, it raises the possibility that other coagulation factors might be co-regulated in the opposite direction. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Repeat testing in individuals with borderline or discordant clinical and laboratory findings'.)

●A few case reports have described abnormal platelet aggregation studies in individuals with psychogenic purpura. As an example, a publication from 2005 described a 16-year-old female with menorrhagia and episodic painful bruising on the neck and back [6]. Platelet function studies revealed a prolonged closure time to epinephrine on platelet function assay (PFA)-100 testing and impaired platelet aggregation to adenosine diphosphate (ADP) and collagen. She was treated with estrogen-containing oral contraceptives and desmopressin (DDAVP), and symptoms improved. One year later she was diagnosed with bipolar disorder. Unusual abnormalities of platelet function might underlie some cases of psychogenic purpura and suggest possible directions for future research.

None of these observations fully explain the appearance of purpura in individuals under psychologic or physiologic stress. The contribution of psychiatric disorders to psychogenic purpura also remains unclear.

Histologic findings — Biopsy findings have not clarified the pathophysiology of psychogenic purpura and are rarely helpful in evaluating or treating individuals with unexplained purpura, and we do not perform biopsies in suspected psychogenic purpura unless there is another indication. (See 'Skin testing (generally not recommended)' below.)

However, case reports have described histopathology of skin lesions, which may help explain the underlying process. Biopsies have shown intradermal hemorrhage and various degrees of edema with scant perivascular lymphocytic infiltrates [7]. In the Mayo Clinic series, 28 individuals underwent skin biopsy [1]. The most common finding was dermal, epidermal, or subcutaneous hemorrhage. Other patients had signs of nonspecific inflammation and other nonspecific findings. We do not advocate biopsy unless there is an alternative diagnosis that is likely enough to justify a biopsy. (See 'Evaluation' below.)

"Autoerythrocyte sensitization" or immune changes — The phrase "autoerythrocyte sensitization" refers to the mechanism that Gardner and Diamond originally ascribed to psychogenic purpura in which there appeared to be sensitivity to extravasated red blood cells that resulted in a painful ecchymotic area [8]. They performed intradermal injection of autologous blood or washed autologous red blood cells, which resulted in an ecchymotic lesion, whereas injection of control blood (eg, from an allogeneic donor or even animal blood) or saline did not [9-11]. We do not support this mechanism.

This finding has been observed in some subsequent case reports and found to be absent in others. Further testing of immune sensitivity such as searches for autoantibodies or cytotoxic T-lymphocytes directed against the individual's own erythrocytes have never been detected. Thus, the phenomenon of immune sensitivity to autologous erythrocytes remains a theoretical idea with limited to no evidence of an actual causal relationship.

One case report described painful ecchymoses in a 22-year-old female who had the symptoms for approximately seven years and experienced an exacerbation when a copper-containing intrauterine device (IUD) was placed [12]. There was a strong family history of atopy. Immunoglobulin (Ig) E levels were increased with her initial episode of purpura. After the IUD was placed, she had frequent episodes of purpura, and intradermal injection of her own blood caused an immediate pruritic wheal and flare, followed by pain after six hours, followed by a bruise that resembled spontaneous bruises. Bruising ceased after the IUD was removed.

A number of other mechanisms and disorders have been postulated as underlying or exacerbating psychogenic purpura. However, because these have been reported infrequently, their relationship to psychogenic purpura may be completely coincidental. It is also possible that the symptoms in these individuals were due to an underlying bleeding disorder rather than psychogenic purpura. (See 'Differential diagnosis' below.)

EPIDEMIOLOGY — The true incidence of psychogenic purpura is unknown, but the condition is thought to be extremely rare. Most available information comes from case reports and very small series. In a review of patients seen at the Mayo Clinic in Rochester, Minnesota from 1976 to 2016, only 76 cases were identified over a period of 40 years [1].

Many clinicians believe psychogenic purpura is most likely to occur in White women, but this has not been rigorously evaluated. The majority of cases reported were in females, although cases in males have been reported. (See 'Patient characteristics' below.)

The typical age is in adolescence or early adulthood. Occasionally, older individuals or children with the condition have been described, such as a series of six girls referred to the Hospital for Sick Children in Ontario, Canada over a five-year period [13].

CLINICAL FINDINGS

Patient characteristics — There is no typical or classic presentation of psychogenic purpura. Many affected individuals are White females; however, the disorder has also been described in children, adolescents, and adult males [6,13-24].

In many cases, affected individuals appear to have severe or acute emotional stress, although this is speculative. Affected individuals may have one or more concomitant psychiatric diagnoses; however, not all affected individuals have a psychiatric condition. (See 'Assessment for psychiatric disorders (all patients)' below.)

Most cases of religious stigmata appear to have occurred in individuals of the Roman Catholic faith. Many are members of Catholic religious orders, and 90 percent are female. Well-known historical cases demonstrating religious stigmata might include Saint Catherine of Siena and Saint Francis of Assisi. In historical yet questionable reports of religious stigmata, the term "religious ecstasy" was used to describe the individual's condition, which some have regarded as an acute psychotic state. Other cases, however, are believed to be "pious fraud" [3,25-27].

Bruising pattern — There is no typical bleeding pattern that characterizes psychogenic purpura. However, certain features such as painful purpura are consistent with the diagnosis, whereas others, such as petechiae, internal bleeding (eg, bleeding occurring in joints, muscles, or visceral organs), or abnormalities of hemostatic testing, are inconsistent with psychogenic purpura. (See 'Painful purpura' below and 'Findings that are inconsistent with the disorder' below.)

Painful purpura — Painful purpuric lesions are the characteristic finding in psychogenic purpura (picture 1). These include ecchymoses (bruises) and/or cutaneous hematomas, with bleeding into the skin or an edematous reaction sufficient to cause induration or a raised area. In many cases, patients have described a sensation of burning, stinging, or pain some hours or up to a day before the appearance of the skin lesions [2,28]. It is also possible that the lesions cause discomfort that is not reported by the patient as pain.

The lesions generally appear as confluent areas of purpura of various sizes; some can be quite large (taking up the entire ventral thigh) whereas others are smaller (similar to the size of a thumbprint). In some cases, skin folds and hair follicles may be spared (picture 2) [29]. The lesions are not at sites of trauma (if they were, they would likely be attributed to injury of self-harm). Pain and swelling may be sufficiently severe to cause immobilization of the affected extremity. However, if the extremity is immobilized due to hemarthrosis, a bleeding disorder is probably present, not psychogenic purpura. (See 'Findings that are inconsistent with the disorder' below.)

Purpura, which is due to extravasated red blood cells, should be distinguished from erythema, which may reflect flushing or hives caused by dilated arterioles and/or localized swelling; flushing and hives are much more clearly linked to emotional discomfort or stress. In some cases, purpura may be accompanied by local inflammation, but inflammation in the absence of purpura is not consistent with the diagnosis. (See "Approach to flushing in adults", section on 'Emotional flushing' and 'Findings that are inconsistent with the disorder' below.)

The bleeding lesions in patients with psychogenic purpura often occur on the extremities rather than the trunk or back; occasionally, they are seen on the head. Some observers have likened them to "religious stigmata," a term that refers to bleeding and/or sensations of pain in locations corresponding to the crucifixion wounds of Jesus Christ [3]. These include wounds from nails in the hands and feet and from a lance in the side. Other reports include wounds to the forehead ("crown of thorns"), tears of blood or sweating blood, wounds to the back as from whipping, or wounds to the shoulder as from bearing a cross [30].

Time-course and natural history — According to some patients, bruises occur either spontaneously or after trauma or surgery at other sites of the body. One patient, for example, had a wrist fracture and a subsequent carpal tunnel syndrome requiring intervention [25]. This was accompanied by severe emotional disturbance; spontaneous bruises began three weeks later on the same arm and hand. While some patients can describe the exact onset of purpura, others are unable to do so even after precise questioning.

The resolution of purpura depends on the size of the lesions, the extent of extravasated blood, and whether new bleeding is superimposed on older bleeding. Typically, resolution takes days to weeks.

The overall time-course and chronicity of the disorder is variable. Some individuals have a single episode that resolves and does not recur [11]. Others have lesions at all times or episodes of recurrent lesions that come and go over the course of years (chronic or relapsing and remitting). In the series from the Mayo Clinic, 28 individuals had follow-up information available in the medical record for a year or more [1]. There was no mention of recurrence in 15 (54 percent) and episodic recurrence in 13 (46 percent). While some patients can describe the exact onset of purpura, others are unable to do so even after precise questioning.

Triggering events — In some case reports, a specific triggering event has been described prior to the first episode of purpura. However, such descriptions are highly subject to recall bias and reporting bias, and the role of a triggering event (psychologic or physical) is not well characterized. In the series of 76 patients with psychogenic purpura from the Mayo Clinic, the following proportions of individuals had a prior event that might have served as a trigger [1]:

●Surgery – 20 (32 percent)

●Trauma – 18 (29 percent)

●Emotional event – 15 (24 percent)

●Abuse – 5 (8 percent)

●Infection – 4 (7 percent)

More than one of these events was present in 11 patients (14 percent). Altogether, patients with an identifiable triggering event accounted for 67 percent of the cohort. In a series of 71 patients from the University Hospitals of Cleveland, preceding injury or surgery was also present in many but not all patients [2].

In 2 of 76 patients in the series from the Mayo Clinic, episodes of purpura appeared to correlate with the menstrual cycle; however, this connection was probably purely coincidental [1].

Preceding or concurrent emotional stress is often described [1,2]. However, the presence or amount of stress cannot be used to support or refute the diagnosis because stress is also a marker of many of the other conditions in the differential diagnosis of psychogenic purpura including abuse and psychiatric disorders. (See 'Terminology and general concepts' above and 'Differential diagnosis' below.)

In one documented case, a large area of "bruising" was noted over the patient's chest when told of hospital discharge that she considered premature; the lesions' area reduced significantly within hours of the patient being told that discharge was postponed [31]. In another case, a 16-year-old began having bruising during a period of time when her mother lived in a different city and the patient was sexually abused by her father [2]. The distinction between traumatic stress and downstream physiologic consequences of stress is not always straightforward. It is also possible for stress to serve as a trigger for self-harming behaviors, factitious disorder, or factitious disorder by proxy. (See 'Differential diagnosis' below and "Factitious disorder imposed on self (Munchausen syndrome)" and "Medical child abuse (Munchausen syndrome by proxy)".)

Findings that are inconsistent with the disorder — While there are no pathognomonic findings of psychogenic purpura, the following findings are inconsistent with the diagnosis:

●Isolated petechiae (very small, red, flat, discrete lesions often occurring in crops in dependent areas), which are typical of thrombocytopenic disorders

●Dental or gum bleeding (so-called "wet purpura")

●Deep tissue bleeding or joint bleeding, indicative of a serious bleeding disorder

●Thrombocytopenia or platelet dysfunction

●Abnormal coagulation testing

●Evidence of von Willebrand disease (VWD) or low VWF, such as a positive family history or abnormal screening tests for VWD

●Retiform or angulated changes such as seen with the purpura that occurs in disseminated intravascular coagulation (DIC) (picture 3)

●Joint deformity

In some of the published case reports, individuals designated as having psychogenic purpura were reported to have more serious bleeding. These individuals may well have had an undiagnosed bleeding disorder or physical trauma (eg, from abuse). As an example, one of the patients in the original publication from Gardner and Diamond was reported to have had an episode of hematemesis and an intracerebral hemorrhage, both types of bleeding that are inconsistent with the rest of the literature on psychogenic purpura [8]. This illustrates the importance of performing a thorough bleeding history and of remaining vigilant for other possible diagnoses.

Level of distress and/or psychiatric comorbidities — Some individuals with psychogenic purpura are distressed by their condition, while others appear unconcerned. As noted separately, however, apparent lack of concern may indicate a self-protective reaction (eg, an attempt on the part of the patient to appear brave, to avoid further questioning, or to avoid a psychiatric diagnosis) rather than true indifference to their condition. (See "Functional neurological symptom disorder (conversion disorder) in adults: Clinical features, assessment, and comorbidity", section on 'History'.)

High levels of stress; stressful life events; complex grief reactions; emotional, physical, or sexual abuse; and/or psychiatric comorbidities are often documented [1,2,32,33]. One case report described a 50-year-old woman with recurrent painful ecchymoses on the extremities that began following the death of her son and returned almost every year at the same time as an anniversary reaction [33]. Others describe a syndrome that begins following an injury or surgical procedure (see 'Triggering events' above). Hypnotic suggestion has been reported to be able to cause and/or to suppress development of the lesions (eg, in patients who previously developed lesions in response to injection of autologous blood, the reaction could be suppressed by hypnosis) [2]. However, apparent emotional distress is absent in some cases [7,34].

Some individuals have been hospitalized for various medical reasons and/or have had multiple surgeries. Malingering and factitious disorder are possible alternative diagnoses in these cases. (See 'Self-Inflicted Injury: Impulse-control disorders, "deception syndromes" (malingering or factitious purpura)' below.)

Pre-existing or concurrent psychiatric conditions are more common than in the general population, although the exact prevalence is challenging to determine. In the series of 76 individuals from the Mayo Clinic, 41 (54 percent) had a current or prior psychiatric diagnosis; the most common diagnosis was depression [1].

Examples of psychiatric conditions associated with psychogenic purpura include:

●Major depression

●Anxiety disorder

●Dissociative disorders

●Personality disorders

●Post-traumatic stress disorder

The connection of these life events and psychiatric conditions with the cutaneous findings are not well understood. (See 'Possible mechanisms' above.)

It is also possible that the evaluation for a bleeding disorder is less extensive in individuals with a major psychiatric condition, leading to an overestimate of the association between psychogenic purpura and psychopathology; however, this bias in evaluation is unlikely to fully account for the high prevalence of psychopathology.

EVALUATION — Psychogenic purpura is a diagnosis of exclusion. There is no specific finding or laboratory test result that can be used to confirm the diagnosis. The evaluation should document the extent and characteristics of the lesions and should reasonably exclude other potential causes of purpura, as discussed in the sections below. When possible, a hematologist, psychiatrist, and/or other specialists (as appropriate) should be involved in the evaluation. (See 'Team approach' below.)

Characteristics of the skin findings — The locations of purpuric lesions need to be clearly and thoroughly noted, with attention to matching up the painful areas with the sites of actual skin bleeding. Pain at sites distinct from the purpuric areas is suggestive of psychogenic purpura.

The patient must be examined thoroughly for scars and signs of other wounds, as well as for venipuncture sites, which may mean that the patient has drawn his or her own blood to be added to body fluids to simulate bleeding. (See "Approach to the adult with a suspected bleeding disorder", section on 'Targeted physical examination'.)

Evaluation for a bleeding disorder

Extent of the evaluation — The extent of the evaluation for an underlying bleeding disorder is a key question for the treating clinician. There is no uniform answer; the amount of testing must be individualized according to the severity of the patient's illness and the likelihood of an underlying bleeding disorder. Consultation with a hematologist is advised for all cases and is especially important in individuals with evidence of serious bleeding.

●As a general rule, individuals with a low bleeding score (ie, <3) from a bleeding assessment tool (BAT) are less likely to have a serious bleeding disorder and may be evaluated with a thorough examination and limited screening tests of hemostasis. The BAT and laboratory tests are discussed separately. (See "Approach to the adult with a suspected bleeding disorder", section on 'Bleeding score' and 'History, examination, and limited laboratory testing (all individuals)' below.)

●In contrast, individuals with a high bleeding score and those who do not fit the more common patterns of bleeding and psychologic stress will require more extensive laboratory testing and/or hematology evaluation. (See 'Additional testing in selected cases' below.)

The presence of a psychiatric diagnosis does not eliminate the possibility of a serious bleeding disorder, and patients should not be denied appropriate testing merely because they appear to have a psychiatric component to their illness.

History, examination, and limited laboratory testing (all individuals) — The initial approach to any patient with a possible bleeding disorder consists of a complete bleeding history. This includes questions about the use (or misuse) of medications that might be associated with bleeding, such as those summarized in the table and discussed below. (See 'Medications associated with bleeding or bruising' below.)

A bleeding assessment tool from the International Society Thrombosis and Haemostasis (ISTH) that lists appropriate questions is available online at https://bleedingscore.certe.nl/.

It is also reasonable to determine whether the patient has access to medications with a high risk of bleeding (eg, warfarin, heparin products, direct thrombin inhibitors, or factor Xa inhibitors). These may be prescribed for other individuals (eg, family members, patients [if a health care provider]). Certain over-the-counter supplements may also be associated with bleeding, such as ginkgo biloba.

A dietary history may uncover unusual dietary practices that could be associated with increased bleeding risk. Examples are vitamin C deficiency due to absence of fruits and vegetables in the diet and vitamin K deficiency due to severe malnutrition. However, the typical bleeding pattern in vitamin C deficiency is perifollicular hemorrhage and gum bleeding; vitamin K deficiency may cause a range of findings from easy bruising to more severe bleeding. (See "Healthy diet in adults".)

The family history and bleeding with challenges such as menses, deliveries, tooth extractions, and surgeries are also highly relevant. Experts stress the importance of using a bleeding assessment tool (bleeding score) to document the bleeding history (or lack thereof) [1]. Additional details about the bleeding history and the use of a bleeding score are presented separately. (See "Approach to the adult with a suspected bleeding disorder", section on 'Patient history'.)

In the review of 76 cases of psychogenic purpura from the Mayo Clinic, a retrospective calculation of bleeding score was done using information from the medical records [1]. The bleeding score was <3 in over 90 percent of the individuals with psychogenic purpura, confirming the absence of significant bleeding complications in this cohort. In many cases, patients had undergone major surgeries without excess bleeding, further supporting the lack of an underlying bleeding disorder.

The physical examination should also evaluate for other conditions that might be associated with bleeding. These include congenital abnormalities associated with inherited platelet disorders, telangiectasia that could indicate hereditary hemorrhagic telangiectasia, and hyperextensibility that might indicate a connective tissue disorder such as Ehlers-Danlos syndrome. (See "Clinical manifestations and diagnosis of hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu syndrome)" and "Clinical manifestations and diagnosis of Ehlers-Danlos syndromes" and "Inherited platelet function disorders (IPFDs)", section on 'Specific disorders'.)

The following laboratory tests are likely to be appropriate in all cases:

●Complete blood count (CBC) with platelet count and review of platelet morphology.

●Screening tests for hemostasis (prothrombin time [PT], activated partial thromboplastin time [aPTT]).

●Testing of renal and hepatic function, as renal insufficiency can cause uremic bleeding and hepatic insufficiency can cause abnormalities of coagulation.

●Testing for von Willebrand disease (VWD) in the majority of individuals if other testing is negative, and especially in those with a family history of VWD or serious bleeding, those with any evidence of mucosal bleeding (which is a common presenting finding), or those with unexplained thrombocytopenia (even if mild) or a prolonged aPTT (the latter are seen in more severe VWD). For clinicians unfamiliar with this testing, hematologist input is advised. Details of testing are presented separately. (See "Clinical presentation and diagnosis of von Willebrand disease".)

Results of the above-listed laboratory testing should be normal in psychogenic purpura. Any abnormality of this testing suggests that another underlying cause of bleeding must be considered and indicates the need for further testing to identify the cause. However, it is important to note that all of these test results are normally distributed, and some individuals with results just outside the normal range may not in fact have any disorder. In such cases, it may be helpful to repeat the testing, and it is essential to interpret the results in the context of the clinical presentation.

Another exception may be a low hemoglobin and hematocrit, which could indicate dietary iron deficiency or a chronic inflammatory state. However, anemia that might be due to bleeding, including heavy menstrual bleeding, may require further investigation.

For some individuals, this testing is sufficient. Examples include those with a low bleeding score (<3), a single episode of purpura, or obvious psychiatric comorbidities that suggest factitious disorder.

For others, such as those with a higher bleeding score (≥3) and lack of a psychiatric comorbidity, additional testing is likely to be indicated, as discussed below.

Additional testing in selected cases — Additional testing may be appropriate in selected cases in which the history and/or examination suggest a less common cause of bleeding. Not all tests need to be done in all patients, and some individuals do not require any additional testing. We generally tailor the testing to the type of bleeding disorder suspected.

Examples of additional testing that may be appropriate in selected individuals include the following:

●Evaluation for VWD in any individual with a positive family history of VWD, more serious bleeding, or mucosal bleeding who was not previously evaluated or for whom results of earlier testing were borderline. In some cases, more specialized testing may be required to diagnose rare forms of VWD. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Laboratory testing'.)

●Testing of factor VIII or factor IX levels in individuals with a first-degree relative with hemophilia A or B, respectively. Although hemophilia is X-linked and manifested in males, typically with bleeding in joints or muscles, females who are heterozygous for a hemophilia mutation can have low factor levels in some cases and may have bruising. (See "Clinical manifestations and diagnosis of hemophilia", section on 'Bleeding in females/carriers'.)

●Evaluation for platelet function disorders in those with abnormal platelet morphology or evidence of more serious bleeding or mucosal bleeding. This may be done using a platelet function assay (PFA)-100 test or platelet aggregation studies. (See "Platelet function testing".)

●Evaluation for a fibrinolytic disorder or a disorder of fibrin crosslinking (eg, factor XIII deficiency) in an individual with more serious bleeding or mucosal bleeding, especially delayed bleeding after initial hemostasis occurred. This may be done using thromboelastography (TEG). (See "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis".)

●Testing for vitamin C deficiency in individuals with a diet lacking in fruits and vegetables and/or those with other characteristic stigmata of vitamin C deficiency such as follicular hyperkeratosis, gingivitis, malaise, neuropathy, and vasomotor symptoms. (See "Overview of water-soluble vitamins", section on 'Vitamin C (ascorbic acid)'.)

●Testing for vitamin K deficiency in individuals with a limited diet or frequent antibiotic use (or supplementation with vitamin K if deficiency is suspected based on diet or laboratory studies such as a prolonged PT). (See "Overview of vitamin K", section on 'Vitamin K deficiency'.)

●Testing for thrombotic disorders such as antiphospholipid syndrome (APS) or heparin-induced thrombocytopenia (HIT) if the skin lesions appear necrotic (as if from ischemia) rather than purpuric. (See "Diagnosis of antiphospholipid syndrome" and "Clinical presentation and diagnosis of heparin-induced thrombocytopenia".)

●Skin biopsy and/or serologies for individuals with suspected small vessel vasculitis and/or polyarteritis nodosa. (See "Evaluation of adults with cutaneous lesions of vasculitis" and "Clinical manifestations and diagnosis of polyarteritis nodosa in adults", section on 'Skin disease'.)

We do not use the bleeding time to evaluate individuals with purpura. This test is rarely performed and has largely been supplanted by the PFA-100. (See "Approach to the adult with a suspected bleeding disorder", section on 'Platelet function testing'.)

Assessment for psychiatric disorders (all patients) — A detailed psychiatric evaluation is of paramount importance if psychogenic purpura is suspected, with information concerning how the patient has responded to major stressful events both recently and in the past (eg, fetal losses, death in the family, divorce, loss of income). The psychiatric evaluation is important due to the pervasiveness of psychopathologic comorbidities and for evaluating other etiologic possibilities.

Psychiatric evaluation should include consideration of the following, with appropriate assessments as determined by the consulting psychiatrist or psychologist:

●Mood disorders, especially major depressive disorder

●Anxiety disorders, especially generalized anxiety disorder and post-traumatic stress disorder

●Personality disorders

●Somatic symptom and related disorders

●History of skin picking or cutting disorders (see "Skin picking (excoriation) disorder and related disorders")

●Psychosis, especially if the patient has an unusual and fixed idea about etiology such as subdermal parasites (see "Delusional infestation: Epidemiology, clinical presentation, assessment, and diagnosis")

In addition, clinicians should always be attuned to the possibility of:

●Domestic abuse

●Major losses, including deaths and others

●Relatives with medically unexplained symptoms, especially similar ones

In some cases, it may be appropriate to evaluate for factitious disorder (deliberate or intentional provoking or inducing of the lesions by the patient) (see 'Terminology and general concepts' above). This must be done thoughtfully, as it implies falsification of symptoms and deceptive behavior on the part of the patient, and as such, is a barrier to trust between the patient and clinicians. As discussed in more detail separately, communications may be most effective if they focus on information gathering and information sharing. (See "Factitious disorder imposed on self (Munchausen syndrome)", section on 'Discussing the diagnosis' and 'How to communicate the diagnosis' below.)

In inpatient settings, if levels of suspicion for self-injury are sufficiently high, a one-to-one in-room monitor may be needed. The reason for the sitter should be explained honestly to the patient. It is sometimes helpful to frame this practice as a means of covering all diagnostic possibilities with equal diligence.

Skin testing (generally not recommended) — Autoerythrocyte sensitization, a potential immune mechanism for the development of purpura following trauma, was proposed in the initial description of the syndrome [8]. (See '"Autoerythrocyte sensitization" or immune changes' above.)

However, the autoerythrocyte sensitization test has never been critically analyzed, clinically validated, or standardized by any laboratory [1]. Sensitivity and specificity of the test are unknown but expected to be low. In the series of 76 patients from the Mayo Clinic, 21 had results of autoerythrocyte skin testing reported, and results were very mixed [1]. The testing was positive in eight, negative in eight, and inconclusive in five. In an earlier series of 71 patients from University Hospitals of Cleveland, the authors reported that the proportion of positive tests declined after they began to lose confidence in the test's validity, perhaps reflecting the patients' inference that the test results were not well correlated with the syndrome [2].

We do not use this test, and we recommend that it not be used in the evaluation. Other means of making the diagnosis of psychogenic purpura such as those discussed above, all of which depend on clinical judgment, should be used.

The test involves injection of approximately 0.1 mL of the patient's blood and a control (animal blood, blood from another individual, or saline) into the patient's skin (intradermally) and observation for the development of purpura around the injected autologous blood but not the control. In principle, this is similar to the intradermal method of allergy testing (see "Overview of skin testing for IgE-mediated allergic disease", section on 'Intradermal method'). However, unlike with allergy testing, this autoerythrocyte testing lacks clinical validity. Further, the procedure could cause pain and/or exposure to infectious agents. Because the injected substances are not standardized, quality controls are not available.

We also do not advocate skin biopsy unless it is justified to evaluate the possibility of another condition for which skin biopsy aids in the diagnosis (eg, if small vessel vasculitis or polyarteritis nodosa are suspected). The utility of biopsy in confirming or excluding findings characteristic of psychogenic purpura is low. (See 'Histologic findings' above.)

If indicated, the evaluation may include attempts to analyze the skin lesions for evidence of self-injection. For individuals with bleeding from other sites, samples may be obtained and tested to determine whether the blood is of human origin and whether it comes from the patient (eg, by comparing red blood cell antigens). This is most likely to apply to patients who visit multiple facilities, those whose symptoms worsen when discharge from the hospital is imminent, or those who worsen without explanation. (See "Factitious disorder imposed on self (Munchausen syndrome)".)

DIAGNOSIS — As noted above, the diagnosis of psychogenic purpura is one of exclusion; no confirmatory finding or test result exists. The extent of testing to eliminate other possible diagnoses such as those summarized below requires clinical judgment of the medical team to decide with reasonable confidence that another cause of purpura cannot be identified. (See 'Differential diagnosis' below and 'Evaluation' above.)

The diagnosis should generally only be made in a patient with purpura that is not due to an abnormality of platelet number, platelet function, coagulation or fibrinolytic abnormalities, skin or vascular disorders, or deliberate falsification of symptoms. Input from the consulting hematologist and psychiatrist in confirming the absence of hematologic or psychiatric disorders is necessary. If this input is unavailable, at a minimum the probable exclusion of these disorders should be explicitly stated.

As with other diagnoses of exclusion, the assessment that other possible diagnoses have been eliminated is never final, and openness to periodic re-evaluation is reasonable, especially when it involves re-evaluating the patient history or incorporating important new information. However, relentless seeking of more testing is not required. (See 'Primary care follow-up' below.)

DIFFERENTIAL DIAGNOSIS — As noted above, psychogenic purpura is a diagnosis of exclusion, and a number of medical and psychiatric disorders are generally eliminated before the diagnosis of psychogenic purpura is assigned. The following summarizes general classes of disorders that should be considered. The intensity of the evaluation for each of these classes of disorders is individualized according to patient and disease characteristics.

Bleeding disorders — Bleeding disorders include abnormalities of platelet function, clotting factors, and the fibrinolytic system that resolves clots. Before a diagnosis of psychogenic purpura is made, the possibility of these conditions should be considered and appropriate evaluations conducted to eliminate them. The intensity of the evaluation is tailored to the patient's presentation, as discussed above (see 'Evaluation' above). In general, diagnosis of one of these conditions is sufficient to exclude the diagnosis of psychogenic purpura.

Like psychogenic purpura, individuals with bleeding disorders may report a history of surgery or trauma (a bleeding challenge) prior to the initial bleeding episode, and bleeding may not become clinically apparent until adolescence or adulthood. Like some bleeding disorders, initial tests of hemostasis (prothrombin time [PT], activated partial thromboplastin time [aPTT], and platelet count) may be normal in some cases.

Unlike psychogenic purpura, bleeding disorders generally cause bleeding with trauma or invasive procedures, and spontaneous serious bleeding (mucosal, gastrointestinal, joint, or deep tissue bleeding) can occur. In bleeding disorders, laboratory testing eventually reveals the type of hemostatic defect, even if initial screening tests were unrevealing. In inherited bleeding disorders, the family history may be positive, and in acquired bleeding disorders, there is often an underlying disease responsible (eg, rheumatic, connective tissue, or malignant disorder).

Details of specific bleeding disorders and the testing that identifies them are summarized above and outlined in more detail in separate topic reviews. (See "Approach to the adult with a suspected bleeding disorder", section on 'Laboratory evaluation' and "Clinical presentation and diagnosis of von Willebrand disease" and "Inherited platelet function disorders (IPFDs)" and "Disorders of fibrinogen" and "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis" and "Rare inherited coagulation disorders" and "Acquired hemophilia A (and other acquired coagulation factor inhibitors)".)

Medications associated with bleeding or bruising — A number of medications can increase the risk of clinically obvious bleeding or convert a small area of purpura to a much larger lesion. These are summarized in the table (table 1) and include the following:

●Antiplatelet agents (eg, aspirin, nonsteroidal antiinflammatory drugs [NSAIDs], P2Y₁₂ receptor blockers, or GP IIb/IIIa inhibitors)

●Anticoagulants (warfarin, direct oral anticoagulants [DOACs], and parenteral anticoagulants [unfractionated heparin, low molecular weight heparin, fondaparinux, and others])

●Selective serotonin reuptake inhibitors (SSRIs)

●Glucocorticoids (systemic or topical)

●Illicit drugs, some of which may be combined with anticoagulants

In some cases, these agents may be omitted from the medical record or the patient's report, either because they are prescribed by another clinician, prescribed for a family member, taken surreptitiously, or obtained over the counter without a prescription and thus not considered to be "medications." As noted above, health care workers and pharmacists may have access to these medications. (See 'History, examination, and limited laboratory testing (all individuals)' above.)

Antiplatelet agents are not detected by routine laboratory testing, although tests of platelet function may be abnormal. (See "Platelet function testing", section on 'Caveats with testing'.)

Some of the anticoagulants will cause abnormalities in clotting tests, but lack of abnormalities does not completely eliminate the possibility of anticoagulant use, as some of the medications have short half-lives. (See "Warfarin and other VKAs: Dosing and adverse effects" and "Heparin and LMW heparin: Dosing and adverse effects" and "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects".)

SSRIs have been reported to inhibit platelet aggregation by decreasing platelet serotonin concentrations [35,36]. Concomitant use of an antiplatelet agent or anticoagulant may aggravate these SSRI-related adverse events [37-40]. In one case, ecchymoses appearing after SSRI treatment were successfully treated with vitamin C [41]. (See "Selective serotonin reuptake inhibitors: Pharmacology, administration, and side effects", section on 'Bleeding'.)

Glucocorticoid-induced purpura typically affects the sun-exposed areas of the dorsum of the hand and forearm and is not accompanied by palpable swelling or pain. (See "Major adverse effects of systemic glucocorticoids", section on 'Dermatologic effects and appearance'.)

Illicit drugs may have effects on hemostasis either via direct effects of the drug or from substances that are combined with the drug, such as vitamin K antagonist poisons in cannabinoids. (See "Synthetic cannabinoids: Acute intoxication", section on 'Life-threatening coagulopathy (brodifacoum adulteration)'.)

Skin, connective tissue, and vascular disorders — The vascular purpuras include severe vitamin C deficiency, structural vascular abnormalities (eg, hereditary hemorrhagic telangiectasia [HHT]), hereditary disorders of connective tissue (eg, Ehlers-Danlos syndrome, osteogenesis imperfecta), small vessel vasculitis, and purpura associated with paraproteins or amyloid involvement of cutaneous vessels.

In most cases, routine history, physical examination, and laboratory testing will be sufficient to suspect these diagnoses. (See "Overview of water-soluble vitamins", section on 'Vitamin C (ascorbic acid)' and "Clinical manifestations and diagnosis of hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu syndrome)" and "Clinical manifestations and diagnosis of hypermobile Ehlers-Danlos syndrome and hypermobility spectrum disorder" and "Osteogenesis imperfecta: An overview" and "Overview of amyloidosis", section on 'Skin manifestations'.)

If a systemic vasculitis is suspected, an autoantibody screen including anti-nuclear antibodies (ANA) and anti-mitochondrial antibodies (AMA) as well as a skin biopsy may be useful.

Occasionally, patients with the antiphospholipid syndrome may present with painful cutaneous lesions showing necrosis, infarctions, and/or ulcerations. (See "Clinical manifestations of antiphospholipid syndrome", section on 'Cutaneous manifestations'.)

Physical abuse — Physical abuse explains the findings in some individuals, and the presence of physical abuse excludes the diagnosis of psychogenic purpura. However, abused patients who develop bruising when their abuser is not present may still warrant the diagnosis of psychogenic purpura; in a series of 71 patients from University hospitals in Cleveland, some patients with psychogenic purpura who had been abused by spouses or parents also developed purpuric lesions when the abuser was not present [2].

Many individuals who are abused do not disclose this on initial questioning. Reasons for this lack of disclosure, many self-protective, and approaches to care for the individual, are discussed in detail separately. (See "Intimate partner violence: Diagnosis and screening" and "Intimate partner violence: Intervention and patient management".)

Self-Inflicted Injury: Impulse-control disorders, "deception syndromes" (malingering or factitious purpura) — Some cases have been reported in which factitious purpura (intentional self-injury to create purpura) was the main diagnostic consideration [42,43]. Self-induced injuries can also occur in impulse disorders or malingering.

●Impulse control – In an impulse-control disorder, a patient would fail to resist impulses to scratch or pinch themselves. If self-presenting, these patients would not typically conceal the nature of their bruising. If presenting at the behest of others (eg, a parent), one should be sure to interview them alone and nonjudgmentally inquire about possible involuntary or "compulsive" self-injury. These patients will often describe the escalation of an urge that is increasingly uncomfortable to resist, followed by engagement in the behavior, which, in turn, is followed by a temporary feeling of relief and, often guilt.

●Deception syndromes – Deception syndromes include malingering and the factitious disorders. In malingering, illness is feigned or exaggerated for an external, tangible incentive (substances, housing/hospitalization, disability benefits). In factitious disorder, the individual or their proxy misrepresents through reporting intentionally causing the abnormality to gain attention, sympathy, treatment, and comfort from medical personnel [44]. (See "Factitious disorder imposed on self (Munchausen syndrome)".)

Except at the margins, malingering and factitious disorder (despite being seen as medicalized lying and a psychiatric disorder, respectively) differ only in terms of presumed differences in what the patient is trying to gain. Thus, we advise providers who suspect deception in an individual with painful purpura focus less on distinguishing between malingering and factitious disorder and instead focus on the patient's motives as only one component of a behavior that ultimately one seeks to redirect in a more productive and less dangerous direction.

Aside from actually catching the patient in the act of creating the lesions, which is exceedingly rare, other components of these types of behaviors include [45]:

•Excessive use of health care services distributed geographically or across health care systems.

•Inconsistencies, implausibilities, hyperbole, or evasiveness in the history.

•Atypical presentations and progressions (including unobserved escalations) of symptoms and signs.

•Reports and findings inconsistent with laboratory and other investigations.

•Unusual patient behaviors that may include improbable knowledge, eagerness for procedures, nonadherence and hostility or unusual obsequiousness, hostility to psychiatric consultation, and pseudological fantastica (dramatically elaborate autobiographical lying).

•Consistently surprising (in the unpleasant sense) longitudinal course and treatment responses.

MANAGEMENT

Team approach — The care of individuals with psychogenic purpura involves a team approach that ensures consistent communication and support for the patient, family, and medical team. In many cases, the team will include a hematologist (at least for the initial diagnostic evaluation and/or any subsequent concerns about a primary bleeding disorder), the psychiatrist, and the primary care clinician.

Supporting the patient — As with any medical condition, trust between the affected individual and the clinicians providing care is a cornerstone of effective therapy. Discussions with the patient about the diagnosis and proposed treatments should be respectful and clearly communicated. Labels such as "difficult patient" should be avoided in the medical record, as they may inadvertently cause other caregivers or the patient to experience a level of hostility that interferes with care.

Involvement of family members may also be advisable, provided there is no concern that the patient is being subjected to abuse or coercion by the family [46].

How to communicate the diagnosis — Communication of similar diagnoses such as psychogenic nonepileptic seizures can be adapted to the communication of the psychogenic purpura diagnosis [47]. Key elements include the following:

●Provide the good news of what has been ruled out

●Reveal the diagnosis, with the proper name and openness about unknown causation

●Reassure the patient that they are not presumed "crazy" or "faking"

●Suggest psychiatric evaluation for comorbidity and possible therapy

●Discourage further testing or inappropriate treatments while emphasizing hope and the potential for symptom resolution

Additional aspects of these types of discussions are presented separately. (See "Functional neurological symptom disorder (conversion disorder) in adults: Treatment", section on 'Presenting the diagnosis'.)

Evaluate and treat psychiatric comorbidities — Treatment of psychiatric conditions is appropriate and has been described as helpful in several case reports and review articles. The following illustrates the range of outcomes with psychiatric intervention:

●In a series of 71 cases of psychogenic purpura from University Hospitals of Cleveland published in 1989, psychiatric symptoms and the need for psychiatric treatment were common [2]. Depression was recorded in 41 (58 percent) and overt sexual problems in 31 (44 percent).

●In the series of 76 cases from the Mayo Clinic published in 2019, the majority of patients had a benign course [1]. Psychiatric treatment was used for those who required it, most commonly for depression (23 of 41 individuals with a psychiatric diagnosis).

●In the series of six children with psychogenic purpura, it was reported that most individuals' symptoms responded well to psychiatric treatment, but that some required "removal of the patient from the home environment," suggesting that abuse was also a component of the clinical picture [13].

Psychiatric treatment may be especially effective in children, adolescents, and those with only a short duration of the disorder [15,48].

Primary care follow-up — The primary care clinician will have the principal responsibility of coordinating the patient's care after discharge from the hospital or completion of the hematologic and psychiatric evaluations.

Although evidence is extremely limited regarding the optimal primary care management of patients with psychogenic purpura, we believe it is reasonable to extrapolate from other syndromes and use the following general approach [49]:

●Work with the patient to alter illness behaviors rather than to eliminate the symptoms completely. This work includes establishing and maintaining a trusting relationship that makes the other components outlined below possible. A focused physical examination at each visit can facilitate this.

●Use frequent scheduled visits to check in with the patient, especially during periods of significant symptomatology. The frequency of visits may be extended as symptoms dissipate. Avoid the temptation to omit these check-ins, as patients with somatic symptom disorders in particular will seek frequent contact in an otherwise more unstructured way.

●Once the diagnosis has been established and is considered accurate and a management plan has been determined, it is advisable to avoid extensive, ongoing consultations with specialists in multiple other areas, as this may result in fragmented care and mixed messages that detract from the primary treatment [50].

As noted above, it is unknown for any individual patient whether the condition will resolve after a single episode, remain chronic, or follow a relapsing and remitting course. (See 'Time-course and natural history' above.)

Supporting the clinicians — We favor a model similar to that used in palliative care, in which clinicians work together to share and discuss the diagnosis and treatment plan. This should include nursing staff, consultants, members of the primary inpatient team, and those who will be responsible for the patient's long-term outpatient care.

Other interventions — There are no specific medications or procedures that have been clearly demonstrated to alter the course of psychogenic purpura, and we do not advocate nonspecific immunotherapies for these individuals.

Specifically, we do not use interventions described in early case reports of psychogenic purpura such as glucocorticoid injections, surgical joint explorations, or splenectomy [8]. These procedures were mostly directed towards immune modification based on the idea that psychogenic purpura was an immune-mediated condition. However, evidence does not exist to support this mechanism or the efficacy of any of these therapies.

Surgical intervention for hemarthrosis should be done if considered necessary. However, if surgical intervention is needed for joint bleeding or damage, the patient probably has a bleeding disorder other than psychogenic purpura.

Case reporting — We highly encourage clinicians who care for a person with psychogenic purpura to document the findings in a case report published in a peer-reviewed journal. Many aspects of the condition are not well understood, including its pathophysiology, typical time-course and natural history of the skin lesions, and efficacy of different treatment approaches. Case reporting is helpful in caring for similar patients and in identifying areas for improvements in care and research.

A national or international registry for individuals with psychogenic purpura is not available.

SUMMARY AND RECOMMENDATIONS

●Definition – Psychogenic purpura (also called Gardner-Diamond syndrome or painful bruising syndrome) describes a poorly understood condition with painful ecchymoses, mostly on the extremities or face, not explained by an underlying bleeding disorder or trauma. Mechanisms are poorly understood and may overlap with functional neurological symptom disorder (conversion disorder) and somatic symptom disorders. (See 'Terminology and general concepts' above and 'Possible mechanisms' above.)

●Prevalence – Psychogenic purpura is rare overall and most common in White women. (See 'Epidemiology' above.)

●Presentation – There is no typical or classic presentation but there is often a triggering event (surgery, trauma, emotional stress, abuse). Painful purpuric lesions (picture 1 and picture 2), often on the extremities, may be preceded by burning, stinging, or pain. Some individuals have a single episode; others have chronic or recurrent episodes. Emotional stress, abuse, or psychiatric comorbidities are often present. Serious bleeding (deep tissue, joints), other types of bleeding (petechiae, mucosal [wet] purpura), or laboratory abnormalities indicative of altered hemostasis are not consistent with the diagnosis. (See 'Clinical findings' above.)

●Evaluation – Psychogenic purpura is a diagnosis of exclusion with no specific confirmatory finding or laboratory result. Skin findings should be well documented. (See 'Diagnosis' above and 'Characteristics of the skin findings' above.)

Bleeding disorder evaluation is tailored to the clinical situation. All individuals should have:

•Thorough bleeding history

•Thorough review of diet, medications, and over-the-counter supplements

•Examination for other disorders

•Complete blood count (CBC), platelet count, and morphology review

•Prothrombin time (PT) and activated partial thromboplastin time (aPTT)

•Creatinine

•Liver function tests

•Testing for von Willebrand disease (VWD), especially when other testing is negative and if there is a positive family history, serious bleeding, or mucosal bleeding

Selected individuals may require additional testing. The autoerythrocyte sensitization test should not be used. (See 'Evaluation for a bleeding disorder' above and 'Skin testing (generally not recommended)' above.)

Evaluation for a psychiatric disorder is important, especially major depression, anxiety disorders, personality disorders, abuse, factitious disorder, and others. Physical, sexual, and emotional abuse appear to be overrepresented. (See 'Assessment for psychiatric disorders (all patients)' above.)

●Differential – The differential diagnosis includes bleeding disorders, connective tissue and vascular disorders, physical abuse, and psychiatric disorders with self-harm. (See 'Differential diagnosis' above.)

●Management – Management requires a team approach including treatment of psychiatric comorbidities. The primary care clinician coordinates care, helping the patient to alter illness behaviors, and maintaining a therapeutic alliance that limits excessive consultations and evaluations. We avoid glucocorticoid injections, surgical joint exploration, and splenectomy. We encourage case reporting. (See 'Other interventions' above and 'Team approach' above and 'Case reporting' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Bruce J Dezube, MD, who contributed to earlier versions of this topic review.