From the aClinical Trials Unit, Kaiser Permanente, San Francisco, California, USA
bDivision of Research, Kaiser Permanente, Oakland, California, USA.
Received 7 June, 2007
Accepted 16 June, 2007
Correspondence to Jeffrey Fessel, MD, 2238 Geary Boulevard, San Francisco, CA 94115, USA. Tel: +1 415 833 2854; fax: +1 415 833 8715
In a well-conducted study involving more than 9000 patients and pancreatitis events from 2001 to 2006, the EuroSIDA investigators report in this issue of AIDS a ‘low overall rate of pancreatitis’ (1.27 cases of pancreatitis per 1000 patient-years) and no increased risk of pancreatitis among HIV-infected patients treated with antiretroviral drugs, or among patients who received 2′,3′-dideoxyinosine (didanosine) or 2′,3′-didehydro-3′-deoxythymidine (stavudine) whether singly or in combination. The only risk factors for pancreatitis found after controlling for confounding were high viral load, lower CD4 cell level, and more advanced disease.
Those of us who have been treating HIV-infected patients since the earliest days of the epidemic, and saw much pancreatitis in those early days, will find the EuroSIDA data interesting and quite surprising, particularly in the light of other reports, including one citing data from as recently as 2001, which described rates that are much higher than those from EuroSIDA. Dutta et al.  analysed 321 patients seen in the period 1993–1994; 45 (14%) developed pancreatitis. Among 8451 subjects enrolled in 20 AIDS Clinical Trials Group studies in 1989–1999, Reisler et al.  reported an overall pancreatitis rate that was almost five times that in EuroSIDA, i.e. 6.1 per 1000 person-years. In a separate report, Reisler et al.  saw pancreatitis in 8.5 per 1000 person-years during the later period, 1996–2001.
As the EuroSIDA investigators mention, previous studies used various methods applied to a variety of patients and report a range of rates. It is unfortunate, however, that the EuroSIDA data do not extend further back in time to show whether their current low rates reflect a decline from previous years or whether their rates have always been low.
The EuroSIDA findings led us to examine our own experience with a large group of HIV-infected individuals treated in the facilities of the Kaiser Permanente Medical Care Program in Northern California during the years 1996–2006, when the number of HIV patients in care ranged between 4000 and 6000 annually. We looked to see whether the data for 2001–2006 show a similarly low annual incidence of pancreatitis as reported by EuroSIDA, and whether there was a decline in these later years (2001–2006) compared with earlier years (1996–2000). We counted as incident pancreatitis events (‘presumptive’ by EuroSIDA standards) the first occurrences of plasma lipase greater than four times the upper limit of normal, amylase greater than six times the upper limit of normal (this level of amylase is unlikely to represent a non-specific increase), or a diagnosis of pancreatitis as captured in the electronic medical record. Figure 1 shows that the annual incidence of pancreatitis according to these criteria was approximately five times that seen by EuroSIDA, and approximately the same as occurred in the 8451 patients reported by Reisler et al. . Furthermore, we observed very little, if any, decline in incidence over the 11-year period.
Equally puzzling in the EuroSIDA report (and something we have not yet analysed in our own data) is the absence of a role for didanosine or stavudine in the etiology of pancreatitis despite the fact that 43% of the EuroSIDA patients used these drugs either singly or in combination at baseline. Pancreatitis was recognized as a complication of didanosine therapy even in the phase 1 studies and its occurrence was clearly related to dosage: whereas only 6.3% of those receiving less than 500 mg/day developed pancreatitis, this increased to 13.3% of those receiving 500–750 mg/day, and rose to 50% of those taking more than 750 mg/day . Another early study of 51 men with AIDS treated with didanosine 10–12 mg/kg per day showed clinical pancreatitis in 12 and asymptomatic elevations of amylase and lipase in another 10 men . Pancreatitis caused by stavudine is also a well-established phenomenon. Among 2163 patients receiving one or both of didanosine and stavudine, the rate was 10 per 1000 person-years for stavudine, eight per 1000 person-years for didanosine and 60 per 1000 person-years for stavudine combined with didanosine .
What might account for the four to fivefold lower rate observed at EuroSIDA sites compared with north America, and for the absence of an association with exposure to didanosine and stavudine? These are intriguing questions that deserve further study. Is the lower incidence of pancreatitis an issue of different drugs besides antiretroviral agents that were used and might exert synergistic or antagonistic effects with antiretroviral therapy vis a vis harm or protection of the pancreas? For example, pentamidine causes pancreatitis and was widely used in inhalation therapy as prophylaxis against pneumocystis pneumonia before the efficacy of sulfamethoxazole/trimethoprim was recognized. If this were the explanation, there should be far fewer cases in the years since 2000 than in earlier years, but Fig. 1 does not show this difference nor did the two reports of Reisler and colleagues [2,3]. Is there a difference in the prevalence of gallstones in the two continents that might be reflected in different rates of pancreatitis? Is the form of alcohol consumed by Europeans different from that which north Americans drink? Does the north American diet cause more hypertriglyceridemia than the European diet?
As regards the absent association of didanosine and stavudine with pancreatitis in the EuroSIDA studies, these drugs are used less in recent years because of their several toxicities. Our own findings show little if any lowering of the incidence of pancreatitis in those recent years (Fig. 1). Concomitant use of methadone reduced the measured areas under the time-concentration curve of didanosine by 63% and of stavudine by 25% ; was methadone more widely used by European than north American patients? Were more of the north Americans than the Europeans receiving treatment for hepatitis C? Among 22 patients infected with HIV and treated for hepatitis C with interferon and ribavirin, five developed pancreatitis; four of these were taking both didanosine and stavudine, one was taking stavudine without didanosine .
In brief, the EuroSIDA observations of a low rate of pancreatitis and a minimal role for didanosine and stavudine are discordant with experience from north America. The reasons for these unexpected findings are unclear and require further study.
Conflicts of interest: None.
1. Dutta SK, Ting CD, Lai LL. Study of prevalence, severity, and etiological factors associated with acute pancreatitis in patients infected with human immunodeficiency virus. Am J Gastroenterol 1997; 92:2044–2048.
2. Reisler RB, Murphy RL, Redfield RR, Parker RA. Incidence of pancreatitis in HIV-1-infected individuals enrolled in 20 adult AIDS Clinical Trials Group studies: lessons learned. J Acquir Immune Defic Syndr 2005; 39:159–166.
3. Reisler RB, Han C, Burman WJ, Tedaldi EM, Neaton JD. Grade 4 events are as important as AIDS events in the era of HAART. J Acquir Immune Defic Syndr 2003; 34:379–386.
4. Grasela TH, Walewander CA, Beltangady M, Knupp CA, Martin RR, Dunkle LM, et al
. Analysis of risk factors asociated with the development of pancreatitis in phase 1 patients with AIDS or AIDS-related complex receiving didanosine. J Infect Dis 1994; 169:1250–1255.
5. Maxson CJ, Greenfield SM, Turner JL. Acute pancreatitis as a common complication of 2′,3′-dideoxyinosine therapy in the acquired immunodeficiency syndrome. Am J Gastroenterol 1992; 87:708–713.
6. Moore RD, Keruly JC, Chaisson RE. Incidence of pancreatitis in HIV-infected patients on nucleoside RTIs. In: 8th Conference on Retroviruses and Opportunistic Infections. Chicago, IL, 4–8 February 2001. Abstract no. 620.
7. Rainey PM, Friedland G, McKance-Katz EF, Andrews L, Mitchell SM, Charles C, Jatlow P. Interaction of methadone with didanosine and stavudine. J Acquir Immune Defic Syndr 2000; 24:214–248.
8. Hester J, Keiser P, Berggren R, Pancreatitis: an emerging complication of HCV treatment in HIV co-infected patients treated with didanosine/stavudine containing regimens. In: 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. 16–19 December 2001. Abstract no. H-739.