Abstract
Objective
No guidelines exist for Pneumocystis jirovecii pneumonia (PJP) prophylaxis in patients with systemic lupus erythematosus (SLE). Limited data are available on incidence of PJP infection and use of PJP prophylaxis. Using a real-world, electronic health record (EHR) cohort, we investigated the frequency of PJP infections as well as patient and provider factors that impacted use and type of PJP prophylaxis.
Methods
In a large, de-identified EHR, we identified possible SLE patients using a previously validated algorithm. PJP ICD-9 or ICD-10-CM billing codes and PJP keywords were used to identify possible PJP cases within this SLE cohort. We assessed for PJP prophylaxis prescribing in all SLE patients using keywords and reviewing medication lists for prophylactic agents. Chart review was used to confirm cases of SLE, PJP, and PJP prophylaxis and to obtain data on demographics, comorbidities, and immunosuppressants.
Results
Of 977 SLE patients, there were only four with confirmed PJP infection. Two of these patients had concurrent Acquired Immunodeficiency Syndrome, and none were on prophylaxis. Of 977 SLE patients, 132 (14%) were prescribed PJP prophylaxis. Of 617 SLE patients ever prescribed immunosuppressants, 128 (21%) were prescribed PJP prophylaxis. Sulfonamides were the most common prophylaxis prescribed (69%), and possible adverse events were documented in 22 out of 117 instances of being placed on a sulfonamide. Patients of younger age, Black race, nephritis, and renal transplant, and on chronic glucocorticoids were all more likely to have PJP prophylaxis prescribed. Patients who were on transplant induction medications, calcineurin/mTOR inhibitors, cyclophosphamide, and mycophenolate mofetil all were more likely to be prescribed PJP prophylaxis compared to other immunosuppressants.
Conclusion
PJP is a rare diagnosis among SLE patients, and prior studies may even overestimate its prevalence. PJP prophylaxis was less common in our cohort than previously described. Adverse events related to sulfonamides used for PJP prophylaxis were relatively rare with lower rates than previously reported. Our study demonstrates real-world PJP prophylaxis prescribing patterns in a large cohort of SLE patients.
Keywords: Systemic lupus erythematosus, Pneumocystis jirovecii, Prophylaxis, Sulfonamide, Electronic health record
Introduction
Pneumocystis jirovecii pneumonia (PJP) is an important opportunistic infection in patients with systemic lupus erythematosus (SLE) with hospitalization rates between 2 to 12 per 10,000 SLE hospital admissions [1,2] and a mortality rate of 50% [3]. Antimicrobial prophylaxis against PJP, specifically trimethoprim-sulfamethoxazole (TMP-SMX), lowers the incidence and mortality of PJP in patients with rheumatic diseases by over 90% [4,5]. Despite proven efficacy, it is not known whether the risks of PJP prophylaxis outweigh its benefits [6], with the relative infrequency of PJP in SLE and that TMP-SMX may cause adverse reactions or SLE flares [7-11].
Additionally, no formal recommendations exist to guide PJP prophylaxis for SLE patients. There are, however, a variety of individual perspectives on when PJP prophylaxis may be appropriate. These have included suggestions based on the specific immunosuppressants used, the underlying autoimmune disease, lymphocyte counts, and even to consider PJP prophylaxis on a case-by-case basis [12-15]. Lack of agreement on prophylaxis indication has thus led to wide differences in reported utilization of PJP prophylaxis [16-18].
Little is known about the real-world utilization of PJP prophylaxis in SLE patients or the patient and provider factors that impact PJP prophylaxis. Risk factors for PJP infection in SLE patients in the Taiwanese population have been reported [19], as have the physician self-reported practice patterns of PJP prophylaxis [16-18]. Current studies are limited by sample size and focus on general rheumatic diseases rather than SLE specifically [20]. We investigated the frequency of PJP infection in a large, established, electronic health record (EHR)-based SLE cohort. We also examined the impact of demographic, provider, and patient characteristics on the use and type of PJP prophylaxis.
Methods
The Synthetic Derivative
After approval from the Institutional Review Board of Vanderbilt University Medical Center (VUMC) (#141222), we used the Synthetic Derivative, a de-identified copy of the EHR at VUMC, which has over 3.2 million subjects with longitudinal follow-up since the late 1980s [21]. The Synthetic Derivative includes demographics, inpatient and outpatient clinical notes, diagnostic and billing codes, medications, lab values, pathology reports, and imaging results from VUMC. The Synthetic Derivative does not contain outside medical records. Information is searchable by keyword query.
Identifying SLE patients
SLE patients were identified using a previously validated algorithm of ≥ 4 counts of the SLE International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9) code of 710.0 and a positive antinuclear antibody (ANA), defined as a titer of ≥ 1:160 [22]. Subjects with ICD-9 codes for systemic sclerosis (710.1) and dermatomyositis (710.3) were excluded. This algorithm has a positive predictive value of 90%, a sensitivity of 86%, a negative predictive value of 98%, and has been externally validated [22,23]. All subjects identified by this algorithm were confirmed on chart review as a SLE case, as diagnosed by a Vanderbilt or external rheumatologist, nephrologist, or dermatologist. Data on our SLE cohort spans from 1989 to 2021. Cases of drug-induced lupus and isolated cutaneous lupus were excluded. Cases included in our analysis represent both pediatric and adult patients.
Identifying possible PJP cases
Within the cohort of SLE patients described above, we then required either a PJP ICD-9 code (136.3), PJP ICD-10 CM code (B59), or PJP keyword. PJP keywords included “PJP”, “Pneumocystis jirovecii”, “pneumocystis”, “PJP pneumonia”, or “pneumocystis pneumonia”. The cohort selection is summarized in Fig. 1.
Fig. 1. Flow chart of possible SLE patients with PJP.
Potential SLE patients with PJP were selected from a de-identified EHR by first applying a validated and published SLE algorithm requiring ≥ 4 SLE ICD-9 (710.0) codes and a positive antinuclear antibody (ANA) titer ≥ 1:160 while excluding ICD-9 codes for systemic sclerosis (710.1) and dermatomyositis (710.3). To this cohort, we then applied PJP keywords and ICD-9 and ICD-10-CM codes. These patients were then all chart reviewed to determine if a PJP case. Of the 371 possible PJP cases, 145 underwent diagnostic testing for PJP resulting in 4 confirmed PJP cases.
Confirmed PJP infection was defined by positive immunofluorescence, polymerase chain reaction, or Gömöri methenamine silver staining on sputum or bronchoalveolar lavage specimen. For SLE patients with confirmed PJP infection, we performed chart review to collect demographics, medication exposure, and laboratory values. Specifically, we assessed for daily glucocorticoid use, in mg of prednisone-equivalent, at time of PJP infection and cumulative dose in the preceding 90 days, presence of non-steroidal immunosuppressants, presence of Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome (HIV/AIDS), and white blood cell and lymphocyte counts at time of PJP infection.
Assessing for PJP prophylaxis
We assessed for PJP prophylaxis on chart review by using the keywords “PJP”, “pneumocystis”, and “prophylaxis” and reviewed each instance to identify if it was associated with use of PJP prophylactic agents. We also reviewed the medication list for any agents used for PJP prophylaxis, including atovaquone, dapsone, pentamidine, primaquine, sulfisoxazole, and TMP-SMX. If any of these medications were ever used, we performed chart review to confirm whether they were used as PJP prophylaxis. Medication use is collected in the Synthetic Derivative through inpatient and outpatient electronic prescribing systems and a validated data-mining program called MedEx, which identifies medications from free text within clinical documents and has been internally and externally validated [24,25].
For all patients in our SLE EHR cohort, we assessed for demographics, presence of renal transplant, and SLE nephritis, defined as either biopsy-proven disease or clinical diagnosis by a rheumatologist or nephrologist. We also assessed for long-term glucocorticoid use, defined as ≥ 90 days of any dose of oral glucocorticoids for an indication of treating rheumatic diseases, disease-modifying anti-rheumatic drugs (DMARDs), and other immunosuppressants based on the medication list and chart review.
For patients in whom PJP prophylaxis was used, we collected which antimicrobial agent was used, glucocorticoid and DMARD use at the start of PJP prophylaxis, specialist prescribing PJP prophylaxis, and inpatient status. Glucocorticoid use was tracked at the date of PJP prophylaxis start and cumulative glucocorticoid use thirty and ninety days prior to this date, recorded as prednisone-equivalent doses. Patients on PJP prophylaxis at multiple, non-consecutive timepoints or with different agents were counted and analyzed by each instance of initiating prophylaxis. Instances of PJP prophylaxis initiated outside of VUMC were included for analysis only when that agent was documented in clinical notes along with start date.
Statistical Analyses
We evaluated for differences in SLE patients who received PJP prophylaxis vs. did not receive PJP prophylaxis using the Mann-Whitney U test for continuous variables and Chi-square or Fisher’s exact test for categorical variables, as there were non-normal distributions in the data. Cases with missing demographic and clinical data were only excluded from analyses using those covariates. We used logistic regression to estimate the association between PJP prophylaxis and age, sex, race, cumulative corticosteroid use, renal transplant, and nephritis. For logistic regression models, odds ratios (ORs) and 95% confidence intervals (95% CIs) are reported. Two-sided p values less than 0.05 were considered significant. Analyses were conducted using R version 4.0.2.
Results
SLE EHR Cohort
Using a validated and published SLE algorithm, we identified 1147 potential SLE cases (21). Of these 1147 possible SLE cases, 977 were confirmed on chart review and have been previously described [26,27]. SLE cases were predominantly female (90%) and White (70%) with 27% Black, 2% Asian, and 1% other races. Hispanic ethnicity was 4%. In the SLE cohort, 28% had SLE nephritis with 77% of these having a renal biopsy documented, 21% with a clinical diagnosis by a rheumatologist or nephrologist, and 2% with missing data on how the diagnosis was made. Of the SLE cohort, 6% received a renal transplant. Demographic characteristics of this cohort are shown in Table 1.
Table 1.
Characteristics of SLE cases.
| SLE cases (n = 977) | |
|---|---|
| Age at first SLE billing code, mean ± SD, Years | 39.3 ± 16.1 |
| Sex, n (%) | |
| Female | 881/976 (90) |
| Male | 95/976 (10) |
| Race | |
| White | 651/926 (70) |
| Black | 247/926 (27) |
| Asian | 22/926 (2) |
| Other | 2/926 (1) |
| Ethnicity | |
| Hispanic | 34/927 (4) |
| SLE nephritis * | |
| Present | 273/975 (28) |
| Kidney transplant recipient | |
| Present | 62/974 (6) |
SLE nephritis defined by renal biopsy or clinical diagnosis by rheumatologist or nephrologist.
PJP Cases
Applying the SLE algorithm with a PJP code or keyword resulted in 371 potential SLE PJP cases. On chart review of these potential SLE PJP cases, only four true cases of PJP were identified (Fig. 1). Mention of PJP in patient charts was often related to differential diagnosis or diagnostic testing. Testing for PJP was conducted via 119 bronchoscopies, 16 sputum samplings, and 10 samplings of non-pulmonary sites. The four cases of PJP in SLE patients are described in Table 2. All four cases were women between 28 and 44 years old. Two were Black and two were White. Two patients had concurrent AIDS with CD4 counts of 17 and 104. Of the two patients without AIDS, both were on daily prednisone. One patient was on azathioprine and prednisone 15 mg daily at time of PJP diagnosis and had cumulatively received 1.12g of prednisone-equivalent glucocorticoids in the preceding 90 days. The other patient was on prednisone 30 mg daily at time of PJP diagnosis and had cumulatively received 3.25g of prednisone-equivalent glucocorticoids in the preceding 90 days. None of these four cases were prescribed PJP prophylaxis. The first patient listed in Table 2 died from PJP and also had concomitant interstitial lung disease.
Table 2.
Characteristics of SLE patients with Pneumocystis pneumonia.
| Age | Sex | Race |
P. jiroveciitestingmodality |
Current daily prednisone dose (mg)* |
Cumulative prednisone dose in previous 90 days (mg)* |
Non-steroid immunosuppressants in previous 90 days* |
HIV/ AIDS1 |
WBC* (cells/ μL) |
Lymphocytes* (cells/μL) |
CD4 count* (cells/ μL) |
|---|---|---|---|---|---|---|---|---|---|---|
| 28 | F | Black | PCR** | 15 | 1,120 | Azathioprine | No | 6300 | 310 | N/A |
| 31 | F | White | GMS*** | 30 | 3,250 | None | No | 2800 | 120 | N/A |
| 43 | F | White | GMS*** | 0 | 0 | None | Yes | 1100 | N/A | 17 |
| 44 | F | Black | GMS*** | 0 | 0 | None | Yes | 3300 | 550 | 104 |
N/A = not applicable, WBC = White blood cell, HIV/AIDS = Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome
Timing in relation to diagnosis of Pneumocystis jirovecii pneumonia diagnosis
Polymerase chain reaction
Gömöri methenamine silver
We applied the same search strategy we used in our SLE cohort to the entire source population to estimate PJP frequency. This search strategy resulted in 19,541 possible PJP cases with an estimation of 10,553 true PJP cases from 1989 to 2022 or approximately 0.3% (10,553/3,400,000) of our source population, similar to the 0.3% (4/1174) for our SLE cohort.
PJP Prophylaxis
In addition to identifying cases of PJP in our SLE EHR cohort, we examined for rates of prescribing PJP prophylaxis. Of 977 SLE patients, 132 (14%) were prescribed PJP prophylaxis. There were 170 unique instances of PJP prophylaxis due to some patients being prescribed prophylaxis at multiple timepoints. Of the 617 SLE patients who had ever been prescribed immunosuppressants, 128 (21%) were prescribed PJP prophylaxis. We compared SLE patients who received PJP prophylaxis vs. patients who did not receive prophylaxis based on demographics and disease characteristics (Table 3). SLE patients who received PJP prophylaxis were more likely to be younger based on age at first SLE billing code (p < 0.001) and age at time of analysis (p < 0.001). SLE patients who received PJP prophylaxis were also more likely to be male (p = 0.02), non-White race (p < 0.001), more likely to have SLE nephritis (p < 0.001), and more likely to have received a renal transplant (p < 0.001). PJP prophylaxis was prescribed to 38% (92/245) of patients with a history of SLE nephritis and 79% (49/62) of renal transplant patients.
Table 3.
Comparison of SLE cases prescribed vs. not prescribed PJP prophylaxis.
| SLE Cases (n = 977) | Prescribed PJP Prophylaxis(n = 170)* |
Not Prescribed PJP Prophylaxis(n = 841) |
p value |
|---|---|---|---|
| Current age**, mean ± SD | 43.5 ± 14.4 | 56.8 ± 16.4 | <0.001 |
| Age at first SLE billing code, years, mean ± SD | 28.6 ± 13.5 | 40.7 ± 16.0 | <0.001 |
| Sex, n (%) | |||
| Female | 144/168 (86) | 767/840 (91) | 0.02 |
| Male | 24/168 (14) | 73/840 (9) | |
| Race | |||
| White | 73/167 (44) | 561/792 (71) | <0.001 |
| Black | 82/167 (49) | 186/792 (23) | |
| Asian | 7/167 (4) | 17/792 (2) | |
| Other | 5/167 (3) | 28/792 (4) | |
| Ethnicity | |||
| Hispanic | 7/165 (4) | 28/794 (4) | 0.66 |
| SLE nephritis *** | |||
| Present | 119/170 (70) | 154/839 (18) | <0.001 |
| Renal Transplant recipient | |||
| Present | 64/170 (38) | 13/841 (2) | <0.001 |
N refers to occurrence of PJP prophylaxis with 170 occurrences of PJP prophylaxis to 132 unique SLE cases.
Current age refers to age at time of analysis.
SLE nephritis defined by renal biopsy or clinical diagnosis by rheumatologist or nephrologist.
Models for PJP prophylaxis
We measured the association between receiving PJP prophylaxis and sex, age, race, and SLE nephritis. Unadjusted and adjusted models are shown in Supplemental Table 1. In a fully-adjusted model including each of these covariates, SLE nephritis (OR = 7.07, 95% CI 4.75 – 10.54, p < 0.001), age (OR = 0.96, 95% CI 0.95-0.97, p < 0.001), and Black race (OR = 1.59, 95% CI 1.51 – 1.68, p = 0.03) were all significantly associated with receiving PJP prophylaxis. As there was collinearity between SLE nephritis, renal transplant, and long-term corticosteroid use, we constructed additional models to evaluate these variables. Renal transplant status (OR = 42.5, 95% CI 20.89 – 86.52, p < 0.001) was significantly associated with PJP prophylaxis after adjusting for sex, race, and age (OR = 0.94, 95% CI 0.93 – 0.95, p < 0.001). Similarly, long-term glucocorticoid use (OR = 4.89, 95% CI 2.32 – 10.31, p < 0.001) was associated with PJP prophylaxis after adjusting for sex, age (OR = 0.96, 95% CI 0.94-0.97, p < 0.001), and race (Black vs. White OR = 2.05, 95% CI 1.39 – 3.02, p < 0.001).
Medication Exposure and PJP Prophylaxis
For each instance a patient was prescribed a given immunosuppressant, we performed chart review to determine if PJP prophylaxis was started. Table 4 summarizes PJP prescribing rates based on immunosuppressants. Compared to patients not on each respective medication, those on renal transplant conditioning regimens (i.e. thymoglobulin, alemtuzumab, or basiliximab), calcineurin or mTOR inhibitors, mycophenolate mofetil, and cyclophosphamide were all more likely to receive PJP prophylaxis. Patients on rituximab were not more likely to have PJP prophylaxis prescribed (Table 5). SLE patients were least likely to be prescribed PJP prophylaxis if on leflunomide or methotrexate. For SLE patients prescribed PJP prophylaxis, 74% were on the equivalent of prednisone 20 mg per day or more. The mean corticosteroid dose, expressed in prednisone mg/day, for SLE patients prescribed PJP prophylaxis was 41 ± 58 mg/day with a median of 30 mg/day.
Table 4.
Concomitant medication use and prescribing patterns for Pneumocystis jirovecii pneumonia (PCP) prophylaxis.
| Immunosuppressive Medication | Prescribed PJP prophylaxis*n (%) |
|---|---|
| Renal transplant conditioning regimens** | 32/56 (57) |
| Calcineurin and mTOR inhibitors | 52/119 (44) |
| Cyclophosphamide | 43/168 (26) |
| Mycophenolate mofetil or mycophenolic acid | 78/326 (24) |
| Rituximab | 20/97 (21) |
| Belimumab | 7/72 (10) |
| Azathioprine | 16/305 (5) |
| Other biologic DMARDs*** | 3/74 (4) |
| Methotrexate | 7/219 (3) |
| Leflunomide | 0/37 (0) |
| Glucocorticoids (≥ prednisone 20 mg/day) | 125/170 (74) |
| Current Regimen, n (%) **** | |
| 0 DMARDS with GC | 19/170 (11) |
| 1 DMARD with GC | 52/170 (31) |
| 2+ DMARDs with GCs | 54/170 (32) |
| 1 DMARD without GC | 9/170 (5) |
| 2+ DMARDS without GC | 28/170 (16) |
| None or Unknown | 8/170 (5) |
| Prescribing Provider | |
| Rheumatology | 64/170 (38) |
| Nephrology | 71/170 (42) |
| Pulmonology/Critical Care | 13/170 (8) |
| Hematology/Oncology | 9/170 (5) |
| Infectious Disease | 2/170 (1) |
| Neurology | 1/170 (1) |
| Unclear | 10/170 (6) |
| Patient Location | |
| Inpatient | 110/170 (65) |
| Outpatient | 50/170 (29) |
| Unknown | 10/170 (6) |
For each instance a patient was started on PJP prophylaxis, we conducted chart review to determine the medications they were on at the time PJP prophylaxis was started.
Medications included thymoglobulin, alemtuzumab, or basiliximab.
Medications included TNF inhibitors, anakinra, and abatacept.
Immunosuppressive regimen at time of PJP prophylaxis start. DMARD (Disease modifying antirheumatic drugs), GC (Glucocorticoids) = Greater than or equal to the equivalent of prednisone 20 mg daily.
Table 5.
Immunosuppressant Use and PJP Prophylaxis.
| Immunosuppressant | Odds Ratio*(95% CI) | p value |
|---|---|---|
| Renal transplant conditioning regimens** | 7.93 (4.53 – 13.87) | < 0.001 |
| Calcineurin or mTOR inhibitors | 5.12 (3.39 – 7.72) | < 0.001 |
| Mycophenolate mofetil | 2.04 (1.46 – 2.86) | < 0.001 |
| Cyclophosphamide | 1.95 (1.31 – 2.89) | < 0.001 |
| Rituximab | 1.33 (0.79 – 2.24) | 0.28 |
Odds ratio calculated based on patients on each respective medication compared to patients not on each medication.
Renal transplant conditioning regimens included thymoglobulin, alemtuzumab, or basiliximab.
At the time of initiating prophylaxis, most patients were prescribed a DMARD(s) in addition to glucocorticoids, specifically the equivalent of ≥ 20 mg daily of prednisone. Of these patients, 31% (52/170) were on a single DMARD and glucocorticoids while 32% (54/170) were on 2 or more DMARDS and glucocorticoids (Table 4). Among patients not on glucocorticoids, patients on 2 or more DMARDS received prophylaxis more often than those on one DMARD. Of SLE patients receiving PJP prophylaxis, 65% (110/170) were admitted to the hospital or inpatient at time of receiving prophylaxis. Of the 170 instances where PJP prophylaxis was started, nephrologists accounted for 42% of these instances and rheumatologists for 38%.
Type of PJP Prophylaxis Prescribed and Adverse Events
Of the patients on PJP prophylaxis, sulfonamides were the most frequently prescribed at 69% (n = 117) followed by dapsone 16% (n = 27), pentamidine 15% (n = 25), and atovaquone 1% (n = 1). No patients were prescribed primaquine. We performed chart review on those prescribed sulfonamides to assess for adverse events attributed to sulfonamides. Of the 117 instances of sulfonamide for PJP prophylaxis, there were 22 possible adverse events that led to discontinuation of the drug. All 22 of these events were with TMP/SMX. In six of these cases, the patient was switched to a different prophylactic agent. Possible adverse events are shown in Supplemental Table 2 with the most frequent events including leukopenia, rash, and hyperkalemia. Only one patient experienced a documented SLE flare with a facial rash that led to discontinuation of prophylactic TMP/SMX. This SLE flare occurred in the context of non-adherence to SLE medications, which was favored to be the cause of the flare.
Discussion
PJP infections were very rare in our large SLE cohort spanning over 30 years with only 4 PJP cases identified. We did not find an increased frequency of PJP in our SLE EHR cohort compared to our source population. We found 371 possible cases of PJP among 1147 SLE patients, of which only 4 had confirmed PJP infections. Neither billing codes nor keywords for PJP were accurate in identifying true PJP cases. Our findings are important because large studies on PJP in patients with SLE and other rheumatic diseases have relied solely on billing codes to identify PJP cases [1,2]. This practice likely overestimates the incidence of PJP, making chart review essential. Another study with multiple autoimmune diseases found similar results with only 21 true PJP cases in 290 patients initially identified [28]. This study also found PJP infections to be rare in SLE, with 4 cases of PJP over a 20-year span at a single, large institution in the Northeast [28].
Only 14% of SLE patients in our cohort were prescribed PJP prophylaxis. Of SLE patients ever prescribed a DMARD, 21% were prescribed PJP prophylaxis. SLE patients who were younger, Black, on chronic steroids, had SLE nephritis, and who had received a renal transplant were all more likely to have PJP prophylaxis prescribed. Among patient factors, renal transplant status was most strongly associated with PJP prophylaxis being prescribed, which is not surprising given the standardized treatment algorithms used in this highly immunosuppressed population. Expectedly, calcineurin and mTOR inhibitors, commonly prescribed to renal transplant recipients, were most significantly associated with prescribing PJP prophylaxis. Further, cyclophosphamide and mycophenolate mofetil, medications used to treat SLE nephritis, were also significantly associated with prescribing PJP prophylaxis.
Prior to our study, real-world data on physician practice patterns of PJP prophylaxis in SLE patients were limited to a small study that combined patients with different rheumatic diseases [20]. PJP prophylaxis was used in 77%, 68%, and 21% of patients on cyclophosphamide, rituximab, and mycophenolate mofetil, respectively, compared to rates in our study of 26%, 21%, and 24%. Such differences may be explained by their grouping of patients with several different rheumatic diseases with higher rates of PJP prophylaxis. This study reported rates of PJP prophylaxis being lowest among SLE patients (26%), compared to idiopathic inflammatory myopathies (50%) and ANCA-associated vasculitis (63%) (20). With different rheumatic diseases carrying differing risks of PJP infection [6], and ANCA-associated vasculitis receiving a formal guideline recommendation for PJP prophylaxis in patients on cyclophosphamide [29], these differences are not surprising.
Larger studies have examined PJP prophylaxis practice patterns in patients with SLE and autoimmune diseases; however, they have summarized physician-reported behaviors rather than analyzing real-world practice patterns [16-18]. Such studies may be limited by bias, with respondents potentially reporting what they expect investigators believe to be appropriate prescribing, rather than their actual prescribing patterns. In two large survey studies, between 50% and 75% of rheumatologists reported routinely using PJP prophylaxis in SLE patients on cyclophosphamide, which is much higher than the 26% seen in our study [16,17]. In the largest survey of rheumatologists on PJP prophylaxis prescribing patterns, which combined patients with different rheumatic diseases, 95% reported using sulfonamides [17], which is much higher than the 69% found in our study or the 73% reported by Schmajuk et al. [20]. This discrepancy may reflect providers’ poor awareness of their own prescribing habits. Alternatively, situation-specific factors such as acute kidney injury and sulfa allergies may limit the use of sulfonamides in clinical practice.
Despite the proven efficacy of TMP-SMX for PJP prophylaxis, prior studies have generated concern over adverse effects and SLE flares with TMP-SMX [7-9]. Of 117 patients who were prescribed a sulfonamide for PJP prophylaxis in our study, 22 experienced possible adverse effects. There was only one patient who discontinued a sulfonamide in the setting of a SLE flare. This patient was non-adherent to their immunosuppressive therapy, which was favored as the reason for the SLE flare rather than the sulfonamide. Previous studies were based on patient-reported survey data while our study used provider data confirmed on chart review [8,10]. Our study examined prophylactic dosing of sulfonamides compared to therapeutic dosing, which may contribute to a lower rate of adverse events in our study.
While our study examines the real-world PJP prophylaxis prescribing patterns in a large SLE cohort, we have several limitations. First, our study was at a single, large academic medical center in the Southeast, which may limit generalizability. Second, due to the fragmented healthcare system, we could not capture cases of PJP prophylaxis or PJP infection that occurred outside of our institution unless documented by a provider in notes. This limitation may underestimate the incidence of PJP infection and rates of PJP prophylaxis in our cohort. Third, as a retrospective study without a control population, our results cannot accurately weigh the risk of PJP infection in the setting of exposure to certain immunosuppressants or other patient factors. Fourth, we were unable to fully assess the effect of glucocorticoid dosing on prescribing PJP prophylaxis, which is an important clinical factor in the decision to prescribe PJP prophylaxis. While the PJP prophylaxis prescription served as a time point at which to assess current and recent glucocorticoid dosing, there was no suitable time point comparator among the patients not receiving prophylaxis. We did look at ever prescribed glucocorticoids of any dose for a period of 90 days or longer to capture “long-term” use and found that such patients were over 4-fold more likely to receive PJP prophylaxis than those not on long-term glucocorticoids.
Conclusion
We used a real-world EHR to evaluate for PJP infections and PJP prophylaxis among patients with SLE. Our findings add to the growing literature demonstrating the rarity of PJP in SLE patients, often despite significant immunosuppression. Real-world use of PJP prophylaxis in SLE patients was much lower than reported by previous, survey-based studies of rheumatologists. Adverse events from prophylactic agents were uncommon and mild. This finding suggests that sulfonamides may not be as problematic in SLE patients as suggested by previous patient survey studies. While our results do not inform recommendations or guidelines for PJP prophylaxis in SLE, our study demonstrates real-world data in a large cohort of SLE patients on what physicians are doing in clinical practice.
Supplementary Material
Acknowledgements
The authors would like to thank Leslie J Crofford, MD for review of the manuscript.
Financial Support
This work was supported by the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases (1K08 AR072757-01, Barnado); National Institutes of Health/National Center for Research Resources (UL1 RR024975, VUMC); National Institutes of Health/National Center for Advancing Translational Sciences (ULTR000445, VUMC); the Rheumatology Research Foundation (K Supplement Award, Barnado), Dermatology Foundation Career Development Award (Career Development Award, Wheless).
Footnotes
Declaration of Competing Interest
None.
Supplementary materials
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.semarthrit.2022.152106.
References
- [1].Murray SG, Schmajuk G, Trupin L, Gensler L, Katz PP, Yelin EH, et al. National lupus hospitalization trends reveal rising rates of herpes zoster and dedines in pneumocystis pneumonia. PLoS One 2016;11:1–11. 10.1371/journal.pone.0144918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Ward MM, Donald F. Pneumocystis carinii pneumonia in patients with connective tissue diseases: The role of hospital experience in diagnosis and mortality. Arthritis Rheum 1999;42:780–9. . [DOI] [PubMed] [Google Scholar]
- [3].Falagas ME, Manta KG, Betsi GI, Pappas G. Infection-related morbidity and mortality in patients with connective tissue diseases: A systematic review. Clin Rheumatol 2007;26:663–70. 10.1007/s10067-006-0441-9. [DOI] [PubMed] [Google Scholar]
- [4].Park JW, Curtis JR, Moon J, Song YW, Kim S, Lee EB. Prophylactic effect of trimethoprim-sulfamethoxazole for pneumocystis pneumonia in patients with rheumatic diseases exposed to prolonged high-dose glucocorticoids. Ann Rheum Dis 2018;77:644–9. 10.1136/annrheumdis-2017-211796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Ganu SA, Mathew AJ, Nadaraj A, Jeyaseelan L, Danda D. Cotrimoxazole prophylaxis prevents major infective episodes in patients with systemic lupus erythematosus on immunosuppressants: A non-concurrent cohort study. Lupus 2021;30:893–900. 10.1177/0961203321995238. [DOI] [PubMed] [Google Scholar]
- [6].Green H, Paul M, Vidal L, Leibovici L. Prophylaxis of Pneumocystis pneumonia in immunocompromised non-HIV-infected patients: Systematic review and meta-analysis of randomized controlled trials. Mayo Clin Proc 2007;82:1052–9. 10.4065/82.9.1052. [DOI] [PubMed] [Google Scholar]
- [7].Izuka S, Yamashita H, Takahashi Y, Kaneko H. Adverse drug reactions to trimethoprim-sulfamethoxazole in systemic lupus erythematosus. Lupus 2021;30:1679–83. 10.1177/09612033211033985. [DOI] [PubMed] [Google Scholar]
- [8].Petri M Antibiotic Allergy in Systemic Lupus Erythematosus: a Case-Control Study. J Rheumatol 1992;19:265–9. [PubMed] [Google Scholar]
- [9].Wolfe RM, Peacock JE. Pneumocystis Pneumonia and the Rheumatologist: Which Patients Are At Risk and How Can PCP Be Prevented? Curr Rheumatol Rep 2017;19. 10.1007/s11926-017-0664-6. [DOI] [PubMed] [Google Scholar]
- [10].Pope J, Jerome D, Fenlon D, Krizova A, Ouimet J. Frequency of adverse drug reactions in patients with systemic lupus erythematosus. J Rheumatol 2003;30:480–4. [PubMed] [Google Scholar]
- [11].Park JW, Curtis JR, Kim MJ, Lee H, Song YW, Lee EB. Pneumocystis pneumonia in patients with rheumatic diseases receiving prolonged, non-high-dose steroids - Clinical implication of primary prophylaxis using trimethoprim-sulfamethoxazole. Arthritis Res Ther 2019;21:1–10. 10.1186/s13075-019-1996-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Feldman CH, Speyer C, Ashby R, Bermas BL, Bhattacharyya S, Chakravarty E, et al. Development of a Set of Lupus-Specific, Ambulatory Care–Sensitive, Potentially Preventable Adverse Conditions: A Delphi Consensus Study. Arthritis Care Res 2021;73:146–57. 10.1002/acr.24095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Winthrop KL, Baddley JW. Pneumocystis and glucocorticoid use: To prophylax or not to prophylax (and when?); That is the question. Ann Rheum Dis 2018;77:631–3. 10.1136/annrheumdis-2017-212588. [DOI] [PubMed] [Google Scholar]
- [14].Khellaf M, Godeau B. Pneumocystose au cours des maladies systemiques. Press Medicale 2009;38:251–9. 10.1016/j.lpm.2008.11.004. [DOI] [PubMed] [Google Scholar]
- [15].Herrou J, De Lastours V. Predictive factors of pneumocystis pneumonia in patients with rheumatic diseases exposed to prolonged high-dose glucocorticoids. Ann Rheum Dis 2018;79:2020. 10.1136/annrheumdis-2018-214718. [DOI] [PubMed] [Google Scholar]
- [16].Gupta D, Zachariah A, Roppelt H, Patel AM, Gruber BL. Prophylactic antibiotic usage for pneumocystis jirovecii pneumonia in patients with systemic lupus erythematosus on cyclophosphamide: A survey of US rheumatologists and the review of literature. J Clin Rheumatol 2008;14:267–72. 10.1097/RHU.0b013e31817a7e30. [DOI] [PubMed] [Google Scholar]
- [17].Cettomai D, Gelber AC, Christopher-Stine L. A survey of rheumatologists’ practice for prescribing pneumocystis prophylaxis. J Rheumatol 2010;37:792–9. 10.3899/jrheum.090843. [DOI] [PubMed] [Google Scholar]
- [18].Wolfe RM, Beekmann SE, Polgreen PM, Winthrop KL, Peacock JE. Practice Patterns of Pneumocystis Pneumonia Prophylaxis in Connective Tissue Diseases: A Survey of Infectious Disease Physicians. Open Forum Infect Dis 2019;6:1–8. 10.1093/ofid/ofz315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Wang W, Lai C, Huang Y, Li T, Tsao Y, Chen W, et al. Pneumocystis jirovecii pneumonia in systemic lupus erythematosus: a nationwide cohort study in Taiwan. Arthritis Care Res (Hoboken) 2021:0–2. 10.1002/acr.24584. [DOI] [PubMed] [Google Scholar]
- [20].Schmajuk G, Jafri K, Evans M, Shiboski S, Gianfrancesco M, Izadi Z, et al. Pneumocystis jirovecii pneumonia (PJP) prophylaxis patterns among patients with rheumatic diseases receiving high-risk immunosuppressant drugs. Semin Arthritis Rheum 2019;48:1087–92. 10.1016/j.semarthrit.2018.10.018.Pneumocystis. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Roden D, Pulley J, Basford M, Bernard G, Clayton E, Balser J, et al. Development of a Large-Scale De-Identified DNA Biobank to Enable Personalized Medicine. Clin Pharmacol Ther 2008;84:362–9. 10.1038/clpt.2008.89. Development. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Barnado A, Casey C, Carroll RJ, Wheless L, Denny JC, Crofford U. Developing Electronic Health Record Algorithms that Accurately Identify Patients with Systemic Lupus Erythematosus April. Arthritis Care Res (Hoboken) 2017;69:687–93. 10.1002/acr.22989.Developing. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Jorge A, Castro VM, Barnado A, Gainer V, Hong C, Cai T, et al. Identifying Lupus Patients in Electronic Health Records: Development and Validation of Machine Learning Algorithms and Application of Rule-Based Algorithms. Semin Arthritis Rheum 2019;49:84–90. 10.1016/j.semarthrit.2019.01.002.Identifying. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Xu H, Stenner SP, Doan S, Johnson KB, Waitman LR, Denny JC. MedEx: A medication information extraction system for clinical narratives. J Am Med Informatics Assoc 2010;17:19–24. 10.1197/jamia.M3378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Jiang M, Wu Y, Shah A, Priyanka P, Denny JC, Xu H. Extracting and standardizing medication information in clinical text - the MedEx-UIMA system. AMIA Jt Summits Transl Sci Proceedings AMIA Jt Summits Transl Sci 2014;2014:37–42. [PMC free article] [PubMed] [Google Scholar]
- [26].Boone JB, Wheless L, Camai A, Tanner SB, Barnado A. Low prevalence of bone mineral density testing in patients with systemic lupus erythematosus and glucocorticoid exposure. Lupus 2021;30:403–11. 10.1016/j.physbeh.2017.03.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Xiong WW, Boone J, Wheless L, Chung CP, Crofford LJ, Barnado A. Real-world electronic health record identifies antimalarial underprescribing in patients with lupus nephritis Wenlu. Lupus 2019;28:977–85. 10.1177/0961203319856088.Real-world. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Mecoli CA, Saylor D, Gelber AC, Christopher-Stine L. Pneumocystis jiroveci pneumonia in rheumatic disease: A 20-year single-centre experience. Clin Exp Rheumatol 2017;35:671–3. [PubMed] [Google Scholar]
- [29].Yates M, Watts RA, Bajema IM, Cid MC, Crestani B, Hauser T, et al. EULAR/ERA-EDTA recommendations for the management of ANCA-associated vasculitis. Ann Rheum Dis 2016;75:1583–94. 10.1136/annrheumdis-2016-209133. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.