Complications Related to Dapsone Use for Pneumocystis Jirovecii Pneumonia Prophylaxis in Solid Organ Transplant Recipients
Dapsone, used for prevention of Pneumocystis jirovecii infections, has been reported to cause hemolytic anemia and methemoglobinemia; its tolerability in solid organ transplant recipients is not well described. We investigated dapsone-related adverse events in patients undergoing solid organ transplantation from 1999 to 2004. Transplant providers identified patients for the investigators who then reviewed the patients' hospital and outpatient records. Sixteen solid organ transplant recipients fit case definitions for dapsone-related hemolytic anemia (n = 11) or methemoglobinemia (n = 5). Median time from event to dapsone discontinuation was 15 days; all patients improved after drug discontinuation. G6PD enzyme activity was normal in all patients whose test results were available. Dapsone may be associated with hemolytic anemia or methemoglobinemia, even with normal G6PD levels. These events are often not promptly recognized, and drug discontinuation is delayed. Dapsone-related hemolytic anemia or methemoglobinemia should be considered in solid organ transplant recipients with unexplained anemia or hypoxia.
Dapsone, a synthetic sulfone that inhibits folate synthesis, was first used in the late 1940s after its discovery as a potent anti-leprosy agent (1,2). In the 1980s, it was shown to be effective prophylaxis against Pneumocystis jirovecii (PCP) infections in patients infected with Human Immunodeficiency Virus (HIV) (3). Dapsone use has increased in this patient population and it is currently recommended as the second line agent for those unable to tolerate trimethoprim sulfamethoxazole (4). Solid organ transplant recipients also require PCP prophylaxis, typically for 1 year following transplantation.
Although frequently used, dapsone has been associated with multiple side effects including hemolytic anemia and methemoglobinemia. These adverse events are usually dose dependent and are more severe in patients with underlying co-morbidities including anemia, cardiopulmonary disease or decreased G6PD enzyme levels (5,6,7). Although adverse events associated with high-dose dapsone use are commonly found in the literature, clinically significant hemolytic anemia and methemoglobinemia related to prophylactic doses are less frequently reported. A literature review of solid organ transplant recipients identified only 3 case reports of methemoglobinemia (4,8) and no cases of hemolytic anemia related to dapsone use.
The paucity of case reports in patients receiving prophylactic dapsone following solid organ transplantation suggests that adverse events are uncommon. Nevertheless, the potential for significant adverse events may be underestimated. In this report, we investigated hemolytic anemia and methemoglobinemia in solid organ transplant recipients receiving prophylactic doses of dapsone from 1999 to 2004 at the Hospital of the University of Pennsylvania.
Patients and Methods
Patients with possible dapsone-related adverse events were initially identified by their transplant care providers. These patients' outpatient and inpatient charts, the transplant database and available laboratory values were then reviewed to determine whether they met case definitions for dapsone-related hemolytic anemia or methemoglobinemia.
A patient was defined as having had dapsone-related hemolytic anemia if the patient had a decrease in hemoglobin while on dapsone, at least one laboratory finding consistent with hemolysis (elevated LDH, elevated indirect bilirubin, decreased haptoglobin and/or peripheral smear significant for schistocytes or burr cells), and an improvement in hemoglobin to at least the patient's pre-dapsone baseline after dapsone discontinuation. When available, hematology consultations were included as supporting or refuting evidence.
A patient was defined as having had dapsone-related methemoglobinemia if the patient developed respiratory difficulty and oxygen desaturation by pulse oximetry while on dapsone, a documented elevation in methemoglobin levels, and either symptom resolution with improvement in pulse oximetry or decrease in methemoglobin levels to normal after dapsone discontinuation.
Patients who met case definitions for dapsone-related adverse events underwent further chart review. Additional data including demographics, past medical history, transplant and rejection history, dapsone use history, results of G6PD enzyme level testing, available information pertinent to the adverse event, treatment(s) given and outcomes were collected. Possible confounding factors for the patients' presenting symptoms were also evaluated. A list of medications known to be associated with hemolytic anemia or methemoglobinemia was compiled and matched against each patient's medication list at the time of the event. Other possible etiologies for the patients' anemia (including iron, B12 or folate deficiency; gastrointestinal bleeding; autoimmune hemolytic anemia and mechanical hemolytic anemia) and respiratory difficulty (including congestive heart failure, pulmonary hypertension, pneumonia and pulmonary embolus) were also searched for and documented.
Twenty-one solid organ transplant recipients with possible dapsone-related adverse events were initially identified by their transplant care providers. After chart review, 16 fit case definitions for dapsone-related hemolytic anemia or methemoglobinemia. They included 10 females and 6 males with a mean age of 57.1 years (range 20–69 years). Fourteen were Caucasian and 2 were African American. Five patients were recipients of kidney transplants, 4 lung transplants, 2 liver transplants, 2 heart transplants, 1 heart/lung transplant, 1 liver/kidney transplant and 1 heart/kidney transplant.
Twelve of 16 patients were started on dapsone because of sulfa drug allergies; allergic reactions included rash (n = 4), unknown/unspecified reaction(s) (n = 4), hives (n = 3) and vomiting (n = 1). Three patients were changed to dapsone because of elevated creatinine levels and 1 patient was changed because of hyperkalemia on trimethoprim sulfamethoxazole. G6PD enzyme activity was normal in all 8 of 16 cases in which tests results were available; 2 were checked during the event.
There were 11 cases of dapsone-related hemolytic anemia (Table 1) and 5 cases of methemoglobinemia (Table 2). Hemolytic anemia occurred at a median duration of 73 days (range 12–761 days) after transplant, 23 days (range 7–735 days) after dapsone was started, and resulted in a mean hemoglobin decrease of 2.5 gm/dL (range 1.2–5.2 gm/dL). Methemoglobinemia occurred at a median duration of 104 days (range 11–1756 days) after transplant. Median dapsone use prior to methemoglobinemia was 48 days (range 7–809 days) and the mean methemoglobin level was 7.2% (range 5.1–10.7%). All patients were receiving dapsone 100 mg daily.
Table 1. Dapsone-related hemolytic anemia
|1999||67||Male||Caucasian||Heart||N/A|| 9.7||7.5||11.4|| 81|| 21|| 50||31|
|2001||59||Female||Caucasian||Liver||N/A||10.1||7.8|| 9.2|| 78|| 21|| 3||19|
|2003||64||Female||Caucasian||Kidney||N/A|| 8.3||6.4|| 8.8|| 39|| 33|| 51||31|
|2003||60||Female||Caucasian||Lung||Normal||10.6||8.3||11.0|| 20|| 10||190||28|
|2003||20||Female||Caucasian||Liver/kidney||N/A|| 9.6||7.3|| 9.5|| 83|| 72|| 9|| 4|
|2003||45||Female||African American||Heart/kidney||Normal2||10.5||5.3||12.7|| 12|| 7|| 3|| 8|
|2004||59||Male||Caucasian||Kidney||Normal||12 ||9.2||13.1|| 24|| 7||142||13|
|2004||67||Male||Caucasian||Liver||Normal|| 9.6||8.4||10.6|| 29|| 23|| 8||12|
|2004||63||Male||African American||Kidney||Normal|| 9.7||7.0||11.2|| 73|| 70|| 5|| 4|
Table 2. Dapsone-related methemoglobinemia
Four patients were hospitalized with a mean length of stay of 19 days (range 3–61 days). Three of the 4 patients were admitted because of hemolytic anemia or difficulty breathing. One patient was admitted because of dehydration secondary to new onset diabetes; during the admission, he was found to have hemolytic anemia. Median time from event recognition to dapsone discontinuation for all patients was 14.5 days (range 3–573 days). In 7 cases, discontinuation occurred greater than 30 days after the symptoms were recognized. The 2 patients with methemoglobinemia underwent extensive cardiopulmonary workup including echocardiograms, CT scans, bronchoscopies and V/Q scans before dapsone discontinuation. Four of the 5 patients with hemolytic anemia had negative radiologic testing or endoscopies/colonoscopies for gastrointestinal bleeding before dapsone discontinuation. The last patient with hemolytic anemia had serial bloodwork and hematology consultations before dapsone cessation greater than 30 days after event recognition.
All patients improved a median of 13 days (range 1–44 days) after dapsone was discontinued; mean hemoglobin increase was 3.1 gm/dL (range 1.4–7.4 gm/dL) and mean methemoglobin concentration decrease was 5.6% (range 4.8–6.5%). Of the patients with hemolytic anemia, 3 received blood transfusions (2 patients had 2 units of packed red blood cells transfused once; 1 patient had 2 units transfused 2 times), 3 were started on recombinant erythropoietin and 1 underwent treatment with both transfusion and erythropoietin. None of the patients were treated with methylene blue or other reducing agents.
Thirteen of 16 patients were receiving tacrolimus at the time of the event, including all patients with methemoglobinemia; 4 of 5 methemoglobinemia patients also received sirolimus. No patients were receiving other medications known to cause hemolytic anemia or methemoglobinemia. These medications were not discontinued.
There were no episodes of rejection associated with these adverse events. There was one death from unrelated causes within 6 months of the event. There were no cases of PCP.
In the last decade, dapsone has become the recommended second line agent for PCP prophylaxis in patients unable to tolerate trimethoprim sulfamethoxazole (8), and its use among immunocompromised patients has increased. However, dapsone use has been complicated by hematologic toxicities, due in part to its N-hydroxylated metabolites (4,9), including dose-related hemolytic anemia and methemoglobinemia (1).
The literature has estimated that 4–13% of patients on prophylactic doses of dapsone develop some degree of hemolytic anemia or methemoglobinemia (3). Previous reports of clinically significant dapsone-related adverse events have typically been associated with higher doses of the medication. However, in our patients, hemolytic anemia and methemoglobinemia occurred with standard prophylactic dosing (dapsone 100 mg daily).
The actual incidence of dapsone-related complications in transplant recipients is unknown and we were not able to estimate it as we could not determine the actual number of patients receiving dapsone during the period of our study. Nevertheless, our data suggest that an increased potential for methemoglobinemia and hemolytic anemia may exist in solid organ transplant recipients. This may be due in part to medication interactions. Dapsone, tacrolimus and sirolimus are all metabolized by the same cytochrome P-450 isoenzyme, CYP3A4 (10,11,12). Perhaps the metabolism of dapsone was affected by concurrent administration of medications, such as tacrolimus and sirolimus, in our patient population, resulting in an increased concentration of N-hydroxylated metabolites. Although dapsone-related complications occurred throughout the first year following transplant, it is interesting to note that many of the dapsone-related adverse events occurred within the first few months after transplant when levels of immunosuppression may be higher. Our patients did not exhibit other risk factors for dapsone-related hemolytic anemia or methemoglobinemia.
Studies have investigated the utility of using cimetidine, a cytochrome P-450 inhibitor, concomitantly with dapsone to decrease methemoglobin formation. Coleman et al. demonstrated that in patients on dapsone for the treatment of dermatitis herpetiformis, cimetidine decreased methemoglobin levels by 27% (13). Cimetidine is not routinely used in our center; consequently, we do not know whether this would have prevented dapsone-related adverse events in our transplant recipients. Dapsone-related adverse events have been previously noted to occur within the first 6 weeks after initiating treatment (14) and our data support a similarly early occurrence. However, there was a significant range of onset in our patients, demonstrating that although most adverse events may occur within 6 weeks after starting dapsone, they can occur at any point during its use.
Despite reports that dapsone-related adverse events are more likely to occur in certain susceptible patients such as those with G6PD deficiency, we found no association with G6PD levels and adverse events in the few patients for whom this information was available. A case report in the literature also noted a case of hemolytic anemia in a patient with normal G6PD enzymes levels undergoing dapsone treatment for tuberculoid leprosy (15). Notably, only 8 of 16 patients started on dapsone had a documented G6PD level in our records. G6PD levels were not part of the routine screening process for transplant candidates and this data was therefore not necessarily accessible in the transplant record. Whether additional patients had normal G6PD levels prior to transplant is unknown. Although we advocate screening any patient considered for dapsone therapy, it is significant that screening patients with G6PD levels will not identify all patients at risk for dapsone-related hemolytic anemia or methemoglobinemia episodes.
In many cases, our patients presented with symptoms related to their methemoglobinemia, despite relatively low levels of methemoglobin. Methemoglobin levels of less than 10–20% are typically asymptomatic, although some present with otherwise asymptomatic cyanosis (5). Patients usually begin developing clinical symptoms such as dizziness, fatigue, dyspnea and headache at methemoglobin levels greater than 20–40%, with concentrations greater than 45–70% associated with coma, arrhythmias and seizures (4,7,8), and those greater than 70% associated with an increased risk of death (5,16). The presenting symptom for all patients who met case definitions for dapsone-related methemoglobinemia was difficulty breathing with oxygen desaturation. Although the methemoglobin levels were not significantly elevated, it has been noted that underlying co-morbidities such as anemia or cardiopulmonary disease can exacerbate symptoms at lower methemoglobin levels. In our study, 4 of the 5 cases of methemoglobinemia had anemia and/or COPD and 3 of the 5 cases occurred in lung transplant recipients.
Dapsone discontinuation was delayed in the majority of patients, reflecting a delay in recognition of dapsone as a potential etiology of the patients' adverse events. It was not uncommon for patients to undergo extensive evaluation of their symptoms prior to medication discontinuation. Given the relative frequency of dapsone associated adverse events in transplant recipients, it is important to consider this possibility early in individuals with unexplained hypoxia, hyperbilirubinemia or hemolysis. It is likely that the increased reporting of cases at our center in 2003 resulted from enhanced physician awareness of this association.
In the absence of prophylaxis, the incidence of PCP in solid organ transplant recipients ranges from 6.8% to 22% (8), necessitating PCP prophylaxis for at least 1 year following transplantation. Although trimethoprim sulfamethoxazole remains the drug of choice for prophylaxis in this population, not all transplant recipients are able to tolerate this medication. Dapsone remains a reasonable second line alternative; however, given the potential for adverse hematologic events, clinicians should closely monitor patients on dapsone, including those with normal G6PD enzyme activity. Given the high incidence of dapsone-related adverse events, dapsone should probably be reserved in solid organ transplant recipients with true intolerance to trimethoprim sulfamethoxazole. The role of other potential alternatives including atovaquone in transplant recipients has not yet been defined.