Epidemiology and pathogenesis of analgesic-related chronic kidney disease

INTRODUCTION — Nonnarcotic analgesic agents have been associated with progressive chronic kidney disease (CKD) since the early 1950s [1,2].

Analgesic nephropathy is a well-described cause of CKD characterized by renal papillary necrosis and chronic interstitial nephritis. It results from prolonged and excessive consumption of analgesic mixtures containing potentially addictive substances in addition to analgesics [3-7].

Analgesic nephropathy has well-defined histologic and radiographic features and used to be an important cause of CKD [3-5]. However, the incidence markedly declined after the withdrawal of phenacetin from the market and with legislation that made combined analgesics available only by prescription.

In addition to phenacetin-containing and non-phenacetin-containing mixtures, individual analgesic agents may also cause kidney damage.

This topic reviews the epidemiology of, risk factors for, and pathogenesis of analgesic nephropathy associated with phenacetin-containing and non-phenacetin-containing mixtures, as well as CKD associated with single analgesic agents. The clinical manifestations and diagnosis of analgesic nephropathy and acute kidney injury (AKI) associated with nonsteroidal antiinflammatory agents (NSAIDs) are presented separately. (See "Clinical manifestations and diagnosis of analgesic nephropathy" and "NSAIDs: Acute kidney injury".)

ANALGESIC NEPHROPATHY ASSOCIATED WITH PHENACETIN-CONTAINING AND NON-PHENACETIN-CONTAINING MIXTURES — Analgesic nephropathy is associated with prolonged and excessive consumption of analgesic mixtures [3-7]. The analgesic mixtures generally contained centrally acting and potentially addictive agents, such as caffeine or codeine, which led to excessive and habitual use of the analgesic drugs (figure 1). This is an important distinguishing feature from CKD associated with single analgesic agents.

Pathogenesis — Analgesic nephropathy is characterized by chronic interstitial nephritis and renal papillary necrosis and calcification [3,8]. It is caused by prolonged ingestion of analgesics, such as aspirin or antipyrine in combination with either phenacetin, paracetamol, or salicylamide. Caffeine, codeine, or barbiturates are then added to make proprietary mixtures [3-7].

The kidney damage induced by analgesics is most prominent in the medulla. The earliest changes consist of prominent thickening of the vasa recta capillaries (capillary sclerosis) and patchy areas of tubular necrosis; similar vascular lesions can be found in the renal pelvis and ureter, suggesting that the primary effect is damage to the vascular endothelial cells [9]. Later changes include areas of papillary necrosis and calcification and secondary cortical injury with focal segmental glomerulosclerosis and interstitial infiltration and fibrosis.

The mechanisms responsible for the kidney injury are incompletely understood. Phenacetin is metabolized to acetaminophen and to reactive intermediates that can injure cells, in part by lipid peroxidation [10]. These metabolites tend to accumulate in the medulla along the medullary osmotic gradient (created by the countercurrent system). As a result, the highest concentrations are seen at the papillary tip, the site of the initial vascular lesions [11].

Aspirin potentiates the nephrotoxicity of both phenacetin and acetaminophen by two possible mechanisms:

●Acetaminophen undergoes oxidative metabolism by prostaglandin H synthase to reactive quinoneimine that is conjugated to glutathione. If acetaminophen is present alone, there is sufficient glutathione generated in the papillae to detoxify the reactive intermediate. However, if acetaminophen is ingested with aspirin, the aspirin is converted to salicylate, which becomes highly concentrated and depletes glutathione in both the cortex and papillae of the kidney. With the cellular glutathione depleted, the reactive metabolite of acetaminophen then produces lipid peroxides and arylation of tissue proteins, ultimately resulting in necrosis and calcification of the papillae (figure 1) [4,11].

●Aspirin and nonsteroidal antiinflammatory drugs (NSAIDs) suppress prostaglandin production by inhibiting cyclooxygenase enzymes. Renal blood flow, particularly within the renal medulla, is highly dependent upon systemic and local production of vasodilatory prostaglandins. Thus, this region, in the setting of combined aspirin and NSAID use, is more prone to ischemic damage.

Epidemiology and risk factors

Overview — Phenacetin-containing combination agents were almost certainly the most important cause of analgesic nephropathy. Analgesic nephropathy was common before phenacetin was removed from over-the-counter analgesics and before legislation was enacted making combined analgesics only available by prescription. Reports suggested that analgesic nephropathy was responsible for 1 to 3 percent of cases of end-stage kidney disease (ESKD) in the United States as a whole, up to 10 percent in areas of North Carolina, 13 to 20 percent in Australia and part of Europe, and up to 50 percent of dialysis patients in two provinces in Belgium [3-5].

Phenacetin-containing combination agents — By the time phenacetin was withdrawn from the market, phenacetin-containing combination analgesics had been implicated as a likely cause of analgesic nephropathy:

●A prospective, 10-year study found that the incidence of a rise in the plasma creatinine concentration was 12 percent in those with heavy use of analgesic mixtures versus only 1.4 percent in nonusers [12].

●A case-control study compared 554 adults with newly diagnosed CKD (plasma creatinine concentration >1.5 mg/dL or 132 micromol/L) with 516 controls [13]. The relative risk for the development of kidney disease was 5.1 times control in those with daily ingestion of phenacetin-containing analgesic mixtures.

●One report prospectively followed 200 active analgesic abusers and a similar number of matched controls for seven years [14]. Kidney function deteriorated in 12 (6.2 percent) of the patients, a relative risk of 6.1 compared with controls. All of the patients were taking analgesic mixtures (including analgesics such as phenacetin, acetaminophen, and aspirin combined with potentially addicting compounds such as codeine and caffeine). Six of 10 patients who were evaluated had evidence of analgesic nephropathy.

●A 20-year prospective of 7275 females showed an increase in the risk of death from kidney disease among phenacetin users compared with nonusers [15].

The nephrotoxicity of phenacetin is dose dependent [8,16]. Decreased concentrating ability or a mild reduction in glomerular filtration rate (GFR) is observed after cumulative phenacetin intake of as little as 1 kg. In comparison, clinically evident kidney disease requires a minimum intake of 2 to 3 kg each of phenacetin and aspirin. This requires six to eight years of a patient ingesting six to eight tablets (or approximately 1 gram) of phenacetin-containing compounds per day. (See "Clinical manifestations and diagnosis of analgesic nephropathy".)

The decrease in availability of phenacetin-containing analgesic mixtures and other combined analgesics led to a reduction in the number of new cases of analgesic nephropathy [3,4,6,17]. As examples:

●In Europe, analgesic nephropathy as a cause of ESKD virtually disappeared in Sweden and Denmark between 1980 and 1990, while the incidence in Switzerland fell from 28 to 12 percent [6,17]. The latter figure undoubtedly includes many patients who already had kidney disease before phenacetin was removed from over-the-counter analgesic preparations.

●In Australia, the decrease in incidence of analgesic nephropathy started in 1985, which was five years after over-the-counter sales of analgesic mixtures were banned [7].

●In an autopsy study of 616 adults in Basle, Switzerland, the rate of analgesic nephropathy fell progressively from 3 percent in 1980 to 0.2 percent in 2000 [18].

Patients with analgesic nephropathy related to phenacetin use continue to be seen, however, because of the slowly progressive course of analgesic nephropathy [19]. In addition, analgesic nephropathy still remains an important, preventable cause of kidney disease in resource-limited countries where combination pills containing phenacetin are still being manufactured and sold.

Non-phenacetin-containing combination agents — It is not clear whether the decrease in incidence of analgesic nephropathy was entirely due to the withdrawal of phenacetin or whether the limitation of non-phenacetin-containing analgesic mixtures played a role [6,11,14,20]. Some data suggest that the decrease in availability of all combination analgesic agents contributed to the decline in analgesic nephropathy [6,7,21]. As an example, by comparison with the data cited above from Australia, in Belgium, where analgesic mixtures (including addictive substances such as caffeine and codeine) were more readily available, analgesic abuse still accounted for over 15 percent of cases of ESKD in 1990 [6]. In the northern part of the country, where analgesic nephropathy was most prevalent, the incidence was 12 percent in 1993 and decreased toward 4.5 percent in 2008 (Belgian Society of Nephrology [NBVN] registry data). In 1999, phenacetin-containing analgesic mixtures were banned, and manufacturers substituted single analgesics for analgesic mixtures. In 2001, analgesic mixtures became available only by prescription.

However, a long-term analysis of these data found no difference in the time trend of the age-specific incidence of ESKD between Australia and Belgium [22].

There have been a number of extensive reviews of the epidemiologic literature concerning the possible relationship between analgesic use and the development of CKD [23-27]. All emphasize that the trials and studies examining this correlation are methodologically flawed. As a result, despite the collective evidence suggesting that chronic analgesic use is associated with CKD, causality, particularly that due to acetaminophen, has not yet been proven [23].

In 2000, in order to address the uncertainty regarding the contributive role of non-phenacetin-containing mixtures, a committee of scientists, selected by regulatory authorities of Germany, Switzerland, and Austria, and the pharmaceutical industry reviewed the available evidence [20]. The committee concluded that there were insufficient data to either support or refute the hypothesis that non-phenacetin-containing analgesic mixtures caused nephropathy.

Given the flaws in previously performed studies, suggested improvements in study design include the following:

●Patients with early-stage disease should be included.

●Explicit knowledge of both the time in which the drug was started and stopped and the amount of drug ingested should be obtained.

●Personal, not telephone, interviews should be performed and must include visual aids to obtain accurate information concerning the type of drug ingested.

●Patients and control individuals should be drawn from the same population source and should be of adequate number to determine statistical significance.

Two subsequent studies examined the association of combined analgesic agents with CKD, with contrasting results:

●In a case-control Swedish study of 926 patients with newly diagnosed CKD (plasma creatinine concentration of 2.8 and 3.4 mg/dL [250 and 300 micromol/L] for females and males, respectively) and 998 control subjects, compared with aspirin use alone, the regular use of aspirin plus acetaminophen was associated with an odds ratio (OR) of 2.2 (95% CI 1.4-3.5) for the development of CKD [28]. Compared with acetaminophen alone, the use of acetaminophen plus aspirin was associated with an OR of 1.6 (95% CI 0.9-2.7) for the development of CKD [28]. Ninety-seven percent of patients reported never using phenacetin, which had been banned by the time the study was performed.

●A population-based case-control study published in 2007 compared incident cases of ESKD among patients <50 years of age- and sex-matched neighborhood controls from 153 dialysis centers in Germany and 17 centers in Australia [27]. Detailed medical and drug histories were obtained by personal interviews from 907 cases and 3622 controls who had never used phenacetin-containing analgesic agents.

Compared with no or very low analgesic use (defined as less than one tablet of any phenacetin-free analgesic compound per month), patients who ever used these compounds or reported high use had no increased risk of ESKD (adjusted OR 0.8, 95% CI 0.7-1.0 and 1.0, 95% CI 0.8-1.3, respectively). There was also no increased ESKD risk among users of just combined products.

Reasons for the disparity between these studies are not known.

Among nearly 1700 healthy females participating in the Nurses' Health Study in the United States, increased acetaminophen use was associated with an increased risk of reduced kidney function [29] (see 'Acetaminophen' below). However, this study found no interaction between aspirin and acetaminophen use, and analgesic mixtures were not studied.

Other large, retrospective studies that examined an association between analgesic use and CKD recruited only a small number of patients who used combination agents and are discussed below [30,31]. (See 'Chronic kidney disease associated with individual analgesic agents' below.)

CHRONIC KIDNEY DISEASE ASSOCIATED WITH INDIVIDUAL ANALGESIC AGENTS — Individual analgesic agents (ie, noncombination agents) may cause CKD, although this issue remains controversial and the associated histologic and clinical features are not as well defined as for analgesic nephropathy associated with phenacetin-containing and non-phenacetin-containing agents.

Phenacetin — The role of phenacetin in analgesic nephropathy was suggested by studies that showed that the decrease in availability of phenacetin-containing analgesic mixtures and other combined analgesics over the past 10 to 15 years led to a reduction in the number of new cases of analgesic nephropathy [3,4,6,17]. In Europe, analgesic nephropathy as a cause of end-stage kidney disease (ESKD) virtually disappeared in Sweden and Denmark between 1980 and 1990, while the incidence in Switzerland fell from 28 to 12 percent [6,17]. The latter figure undoubtedly includes many patients who already had kidney disease before phenacetin was removed from over-the-counter analgesic preparations. In Australia, the decrease in incidence of analgesic nephropathy started in 1985, which was five years after over-the-counter sales of analgesic mixtures were banned [7].

The nephrotoxicity of phenacetin alone has not been assessed in humans, since the drug has never been available as a single (ie, noncombination) agent. Animal studies have demonstrated that phenacetin alone is the least nephrotoxic of many analgesic agents, suggesting that phenacetin is only nephrotoxic when administered in combination with other agents [32].

Acetaminophen — There is suggestive, but not definitive, evidence that chronic, especially daily, acetaminophen (paracetamol) use has dose-dependent, long-term nephrotoxicity [33]. Acetaminophen is the primary metabolite of phenacetin and is widely used as a minor analgesic. Multiple [13,16,28,34], though not all [27,35], small studies have suggested that acetaminophen alone is nephrotoxic, although the risk is less than that of phenacetin-aspirin combinations containing addictive substances [13,16,28,34,35].

The association of acetaminophen with CKD was suggested in a case-control study of patients with ESKD [16]. In this report, there was an association between the cumulative intake of acetaminophen and the relative risk of kidney failure: When compared with a relative risk of 1.0 in patients ingesting <1000 pills during their lifetime, the relative risk was 2.0 if between 1000 and 4999 pills were taken and 2.4 if >5000 pills were taken [16].

However, interpretation of these results is confounded by the possibility that patients with kidney disease are more likely to (1) have symptoms requiring analgesics and (2) preferentially use acetaminophen because they are told to avoid aspirin and nonsteroidal antiinflammatory drugs (NSAIDs).

Other studies that analyzed the risk of CKD associated with acetaminophen included patients with less severe kidney dysfunction and, as a result, may have avoided these confounding issues:

●One report looked at analgesic use in 554 adults presenting with new kidney function impairment (plasma creatinine concentration of 1.5 mg/dL [130 micromol/L]) rather than ESKD [13]. It was estimated that the adjusted odds ratio (OR) for kidney disease was 3.2 in patients taking daily acetaminophen. There was no increase in risk with weekly or less frequent use.

●In the Swedish study cited earlier, which included 926 patients with newly diagnosed CKD (plasma creatinine concentration of 2.8 and 3.4 mg/dL [250 and 300 micromol/L] for females and males, respectively) and 998 control subjects, the use of acetaminophen was associated with an OR of 2.5 (95% CI 1.7-3.6) for the development of CKD [28]. There was a progressive increase in risk with increasing cumulative dose (OR of 1.2, 1.3, and 3.3, with a cumulative dose of 1 to 99 grams, 100 to 499 grams, and >500 grams, respectively).

●The correlation between acetaminophen and CKD was evaluated in over 11,000 initially healthy males enrolled in the Physicians' Health Study [30]. The quantitative exposure to analgesics was determined retrospectively via a questionnaire administered at the end of the 14-year study period. Based upon a multivariate analysis, an increased relative risk of CKD (defined as a creatinine >1.5 mg/dL [133 micromol/L]) was not associated with exposure to ≥2500 pills of acetaminophen (OR 0.83, 95% CI 0.50-1.39).

Some limitations with this study include the inability to extend this conclusion to patients with preexisting kidney disease; reliance upon a retrospective questionnaire; lack of knowledge concerning exact dosing of analgesics; and the use of a single serum creatinine measurement.

●A comparison of baseline (1982) and subsequent serum creatinine measurements (1996) was performed among 4494 initially healthy males enrolled in the previously described Physicians' Health Study [31]. Based upon a multivariate analysis, an increased serum creatinine concentration was not associated with exposure to ≥2500 pills of acetaminophen (OR 1.02, 95% CI 0.55-1.90). Other than the ability to measure the change in kidney function over time, this study has similar limitations to the previous report from the Physicians' Health Study.

●Among nearly 1700 healthy females participating in the Nurses' Health Study in the United States, increased acetaminophen use was associated with an increased risk of reduced kidney function [29]. Compared with those consuming <100 grams of acetaminophen, those consuming >3000 grams of acetaminophen had a multivariate OR of 2.04 (95% CI 1.28-3.24) for a decrease in glomerular filtration rate (GFR) of ≥30 mL/min/1.73 m² over an 11-year period.

Papillary necrosis, similar to that seen with analgesic mixtures, is present in at least some affected patients [34]. This is in contrast to the tubular injury and acute, reversible kidney failure that can be induced by an acute acetaminophen overdose. (See "Acetaminophen (paracetamol) poisoning in adults: Pathophysiology, presentation, and evaluation", section on 'Acute kidney injury (acute renal failure)'.)

Aspirin — In most [16,29-31,35-37], but not all [28,38], studies, the long-term administration of aspirin alone (in therapeutic doses) did not appear to induce kidney injury.

●In the Physician’s Health Study cited previously, among 11,000 initially healthy males, based upon a multivariate analysis, an increased relative risk of CKD (defined as >1.5 mg/dL [133 micromol/L]) with ingestion of ≥2500 pills was not associated with exposure to aspirin (OR 0.98, 95% CI 0.53-1.81) [30]. As discussed above, limitations with this study include the inability to extend this conclusion to patients with preexisting kidney disease; reliance upon a retrospective questionnaire; lack of knowledge concerning exact dosing of analgesics; and the use of a single serum creatinine measurement. (See 'Acetaminophen' above.)

●A comparative evaluation of baseline (1982) and subsequent serum creatinine measurements (1996) among 4494 initially healthy males enrolled in the Physicians' Health Study showed no increased risk of CKD associated with aspirin (OR 0.75, 95% CI 0.35-1.57) [31].

●Among approximately 1697 females participating in the Nurses' Health Study in the United States, lifetime aspirin use was not associated with a change in GFR over an 11-year period (1989 to 2000) [29].

Aspirin may potentiate the toxicity of phenacetin and acetaminophen. (See 'Pathogenesis' above.)

Nonsteroidal antiinflammatory drugs — Exposure to large quantities of NSAIDs can probably induce CKD in some cases, although the percentage of patients affected is small relative to the number of prescriptions written [39,40].

A possible role of NSAIDs in the development of CKD was suggested by the acetaminophen case-control trial cited above, which found an increased incidence of ESKD only in those NSAID users who had taken >5000 pills during their lifetime [16]. Another large case-control study suggested that daily NSAID use for over one year may be associated with an increased risk of an otherwise unexplained and newly diagnosed plasma creatinine concentration >1.5 mg/dL (132 micromol/L) [41]. However, the significance of these findings was undermined by two observations: There was no obvious dose dependence, and the decline in kidney function was essentially confined to males >64 years of age; there was no decline in kidney function in the three other groups in this study [42].

None of these studies determined what type of kidney disease might be present. A third report evaluated 259 patients over a 10-year period, among whom 69 patients satisfied the clinical criteria for analgesic nephropathy (including the presence of papillary necrosis) [43]. Twenty-nine had consumed only NSAIDs (primarily for rheumatologic disease), and varying degrees of kidney function impairment were noted in many of these patients.

Large epidemiologic studies have failed to convincingly demonstrate an association between chronic NSAID use and CKD among initially healthy individuals:

●The correlation between CKD and NSAID use was evaluated in over 11,000 initially healthy males enrolled in the Physicians' Health Study [30]. Based upon a multivariate analysis, an increased relative risk of CKD (defined as >1.5 mg/dL [133 micromol/L]) with ingestion of ≥2500 pills was not associated with exposure to NSAIDs. Limitations of this study are discussed above. (See 'Acetaminophen' above.)

●A comparative evaluation of baseline (1982) and subsequent serum creatinine measurements (1996) among 4494 initially healthy males also showed no association between CKD and ingestion of ≥2500 NSAID pills [31].

●Among nearly 1697 females participating in the Nurses' Health Study in the United States, no correlation was observed between change in GFR over an 11-year period (1989 to 2000) and lifetime use of NSAIDs [29].

NSAIDs are common causes of drug-induced acute interstitial nephritis (AIN). (See "Clinical manifestations and diagnosis of acute interstitial nephritis", section on 'Drugs'.)

Patients who develop NSAID-induced AIN are at risk for CKD from chronic interstitial nephritis. This was suggested by review of 1068 biopsy-proven cases of AIN from 1968 to 1997 [44]. Causes, symptoms, and the clinical and histologic findings were compared between patients with permanent and reversible kidney function impairment after AIN.

While NSAIDs accounted for only 20 percent of the cases of AIN, nearly two out of three patients from the NSAID subgroup were found to have permanent kidney function impairment at follow-up, representing the greatest frequency of any of the drug-induced interstitial nephritis.

Subacute symptoms, prolonged exposure, and analgesic or NSAID use were associated with chronicity. Biopsy findings including tubular atrophy, interstitial granuloma, and pronounced interstitial cell infiltration were predictive of chronicity.

NSAID use may cause progression among patients who have preexisting kidney disease. A retrospective study of 10,184 older adult patients administered NSAIDs examined progression of kidney disease for a mean of 2.75 years [45]. A rapid decrease in estimated GFR (eGFR; defined as a decrease >15 mL/min/1.73 m²) was observed in 13.3 percent of patients. After adjusting for age, sex, and comorbidities, high cumulative NSAID use (>90^th percentile for all subjects) was associated with a 26 percent risk of a rapid decline in eGFR relative to those who did not use NSAIDs (OR 1.26, 95% CI 1.04–1.53). In addition, each 100-unit increase in defined daily NSAID dose was associated with a decrease in eGFR of 0.08 mL/min/1.73 m². A clear linear relationship was noted between cumulative dose and eGFR reduction. The mechanism of NSAID-associated CKD is not known; however, it is possible that recurrent bouts of acute kidney injury (AKI) lead to CKD, or chronic NSAID ingestion may promote unrecognized AIN with associated fibrosis or chronic papillary necrosis [46]. (See "NSAIDs: Acute kidney injury".)

It is important for clinicians to be aware of the association between NSAID use and the progression of CKD. Routine reporting of the eGFR may increase clinician awareness of CKD and decrease NSAID use. This was examined in a retrospective study of NSAID-prescribing patterns among 1522 patients with CKD stage 3, 4, and 5 before and after implementation of eGFR reporting to clinicians [47].

A 10.2 percent reduction in NSAID prescriptions followed eGFR reporting, suggesting that clinicians were aware of the nephrotoxic potential of NSAIDs in CKD patients. From the standpoint of kidney function, eGFR increased from 45.9 to 46.9, 23.9 to 27.1, and 12.4 to 26.4 mL/min/1.73 m² in the CKD stage 3, 4, and 5 patients, respectively, after reporting of the eGFR. This study strongly suggests that eGFR should be considered when prescribing medications to CKD patients.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic kidney disease in adults".)

SUMMARY

●Analgesic nephropathy is characterized by chronic interstitial nephritis and renal papillary necrosis and calcification. It is caused by prolonged and excessive consumption of analgesic combination medications. Individual analgesic agents may also cause chronic kidney disease (CKD), although the associated histologic and clinical features are less well defined than for analgesic nephropathy associated with phenacetin-containing and non-phenacetin-containing mixtures. (See 'Introduction' above.)

●Analgesic nephropathy used to be an important cause of CKD, particularly in Australia, parts of Europe (Belgium, Switzerland), and the United States. The decrease in availability of analgesic-combined medications with or without phenacetin has led to a marked reduction in the number of new cases. However, it still remains an important, preventable cause of kidney disease in resource-limited countries where combination pills are still being manufactured and sold. (See 'Introduction' above and 'Epidemiology and risk factors' above.)

●It is possible that long-term use of individual analgesic agents may also lead to a reduction in glomerular filtration rate (GFR). (See 'Chronic kidney disease associated with individual analgesic agents' above and 'Acetaminophen' above.)

●Aspirin alone (in therapeutic doses) does not appear to induce kidney injury but may potentiate the toxicity of phenacetin and acetaminophen. (See 'Aspirin' above.)

●Although it is clear that nonsteroidal antiinflammatory drugs (NSAIDs) can cause acute kidney injury (AKI), the role of NSAIDs in the development of CKD is uncertain. The exposure to large quantities of NSAIDs can probably induce CKD, although the percentage of patients affected is small relative to the number of prescriptions written and even smaller relative to the number of NSAIDs that are bought over the counter. The effect of long-term use of selective cyclooxygenase (COX)-2 inhibitors on kidney function is unknown. (See 'Nonsteroidal antiinflammatory drugs' above.)