You are on page 1of 4

Downloaded from on March 11, 2018 - Published by group.bmj.

ARD Online First, published on February 19, 2018 as 10.1136/annrheumdis-2017-212588

Pneumocystis and glucocorticoid use: created two groups similar in their like-
lihood to receive prophylaxis based on
disease characteristics and underlying risk
to prophylax or not to prophylax (and factors for PJP (those risk factors beyond
the use of high-dose glucocorticoids, a
when?); that is the question risk that was present in all participants).
Crude incidence among the unprophy-
laxed was significantly higher, and when
Kevin L Winthrop,1 John W Baddley2 controlling for bias through propensity
scores, use of TMP/SMX was associated
with a 93% decrease in incidence of PJP.
Pneumocystis jiroveci is an opportunistic ‘high-dose’ glucocorticoids (>30 mg/day) Only one case occurred in the prophylaxis
fungus with the ability to cause lethal for 4 or more weeks. Within this cohort, group, and this after initial TMP/SMX was
pneumonia in those with advanced immu- they selected patients offered TMP/SMX stopped due to an adverse drug reaction.
nosuppression.1 Fortunately, this outcome prophylaxis and compared their inci- The protection therefore appeared nearly
is preventable with prophylaxis. Unfortu- dence of PJP with the remainder of the complete and those at highest risk were
nately, however, deciding who is immuno- group that did not receive prophylaxis. protected. The benefits of TMP/SMX
suppressed enough to justify prophylaxis There were important underlying differ- cannot be understated, and this study adds
can be a confusing subject, particularly ences between the groups, as one might to others showing similarly high levels of
among rheumatology patients where expect, and it was clear that the treating protective effects within different settings
immunosuppression waxes and wanes physicians had generally chosen to give of immunosuppression.4 So, we know that
based on the use of immunosuppressive TMP/SMX to those they had perceived it works, but what other conclusions can
therapies and the contribution of the at higher risk for PJP. These risk factors we draw from this experience that are
underlying inflammatory disease. Foggy included lymphopaenia, greater glucocor- relevant to the practice of rheumatology?
notions persist regarding who is at risk, ticoid use in the past, concomitant use of
the level of absolute risk where the cyclophosphamide and the presence of
risk-benefit of using trimethoprim/sulfa- dermatomyositis, microscopic polyangiitis When to start prophylaxis?
methoxazole (TMP/SMX) or other (MPA) or granulomatosis polyangiitis. The observations from these researchers
prophylaxis is worthwhile, and when can These factors being more prevalent within confirm the answer is glucocorticoid
prophylaxis be safely stopped.2 The article the prophylaxis group create ‘confounding dose-dependent. This is intuitive, although
by Park et al published in the Annals of by indication’ or ‘channelling bias’ such to our knowledge this has not been shown
Rheumatic Diseases3 sheds light on these that one might expect a higher incidence previously with regard to time to event,
important questions, such that a picture of of PJP in the group receiving prophy- in that patients starting on 30 mg dosing
how to approach this issue clinically might laxis, making it difficult to ascertain any took several months longer on average to
finally be more clear for the practicing protective effect of prophylaxis. This bias develop PJP than those who started 60 mg/
rheumatologist. is the bane of observational and pharma- day. This in part could be related to the
coepidemiological studies, as researchers fact that patients starting higher doses take
struggle to compare ‘apples with apples’ much longer to taper below a threshold
What is the benefit of TMP/SMX
and overcome this bias. In this case, level of risk. They spend longer times at
prophylaxis? the researchers used propensity scores risk. While Park et al’s data suggest you
Given the difficulty studying these ques- to adjust for differences in groups and have longer to make a decision regarding
tions in a randomised controlled trial
fashion, comparative effectiveness studies
such as this one might provide the next
Table 1  Proposed PJP* prophylaxis with glucocorticoid use
best thing. In the article, the authors
address the following questions: in Discontinuation
patients starting high-dose glucocorticoids of prophylaxis at
Prophylaxis at glucocorticoid dose (Y/N)† glucocorticoid dose (Y/N)
and taking them greater than 4  weeks,
what is the risk of pneumocystic jiroveci Underlying disease 15–30 mg >30 mg <15 mg
pneumonia (PJP) (and the risk factors Granulomatosis with Y Y Y‡
for it) and how does risk relate to dose? polyangiitis
Further, they evaluate the efficacy of Microscopic polyangiitis Y Y Y‡
TMP/SMX prophylaxis. To answer these Systemic sclerosis Y§ Y Y
questions, they retrospectively identi- Dermatomyositis/ Y§ Y Y
fied an institutional cohort of rheuma- polymyositis
tology patients in Korea treated with Systemic lupus N Y Y
Division of Infectious Diseases, Oregon Health and Rheumatoid arthritis N Y§ Y
Sciences University, Portland, Oregon, USA *Based on limited data and expert opinion, the authors advocate additional studies to further refine
Division of Infectious Diseases, University of Alabama, recommendations in this area.
Birmingham, Alabama, USA †Requires prolonged glucocorticoids (≥4 weeks).
Correspondence to Dr Kevin L Winthrop, Division of ‡Conditional on <2 additional risk factors at time of discontinuation: baseline lymphopaenia, low CD4 count,
Infectious Diseases, Oregon Health and Sciences cyclophosphamide use, anti-TNF or rituximab use, initial glucocorticoid dose of >60 mg.
University, Portland, Oregon 97239-3098, USA; §Conditional on at least one additional risk factor: baseline lymphopaenia, low CD4 count, cyclophosphamide use,
​ inthrop@​ohsu.e​ du anti-TNF or rituximab use, initial glucocorticoid dose of >60 mg.

Winthrop KL, Baddley JW. Ann Rheum Dis Month 2018 Vol 0 No 0    1
Copyright Article author (or their employer) 2018. Produced by BMJ Publishing Group Ltd (& EULAR) under licence.
Downloaded from on March 11, 2018 - Published by

prophylaxis start in those using 30  mg tolerated TMP/SMX fairly well, with those starting higher doses such 60 mg/day
(versus 60 mg for example), it is clear that approximately 15% of patients devel- where the short-term risk of PJP is much
in any patient where such a dose is envi- oping AEs attributable to TMP/SMX. This greater. This is no matter their under-
sioned for 4 weeks or longer, that prophy- was similar to findings from other anal- lying disease state, although the benefits
laxis should be initiated. yses of rheumatic disease patients using of prophylaxis are greater in those with
TMP/SMX prophylaxis. The authors did higher risk diseases (eg, vasculitis) and
not report what percentage of those using other risk factors. We recommend such
When to stop prophylaxis? TMP/SMX withdrew drug due to adverse prophylaxis should continue until doses
The concept of a risk ‘threshold’ has forever events; however, the incidence of serious are below 15 mg/day, and even at this level
been a ‘holy grail’ type of question, with adverse events attributable to TMP/ if other PJP risk factors such as cyclophos-
initial thresholds drawn at 15 mg or 20 mg SMX was low (n=2 events), such that the phamide use, lymphopaenia or underlying
per day for greater than 3 weeks. These NNTs in order to prevent one PJP case vasculitis are present (table 1). While the
were based on initial case series data that were lower than the NNH with regard to absolute risk of PJP is considered to be low
suggested most cases occur at these dose serious adverse events. Not surprisingly, within rheumatology, this analysis clearly
levels or higher after a prolonged time this benefit-risk scenario varied by disease shows the risk is substantial within certain
period.5 The current study is consistent state where the NNT was markedly lower subgroups of diseases. This is a potentially
with these prior case series and, impor- for the higher risk diseases such as MPA or lethal and preventable infection, and the
tantly, highlights that this threshold with systemic lupus erhythematosus (SLE). Park et al analysis suggests there is little
regard to dose and time is not uniform. There were some important limitations reason to take a risk with a month or more
Three (10%) of the cases within this series to this analysis. Most PJP cases were diag- of high-dose glucocorticoids.
were diagnosed with PJP after tapering to nosed by use of PCR testing on induced
doses below 15 mg/day; however, each case sputum or bronchoalveolar lavage fluid, Handling editor Tore K Kvien
had at least one other risk factor for PJP. We a situation in which it is sometimes diffi- Contributors  KLW and JWB shared all writing
agree with the authors that at this level of cult to distinguish between colonisation responsibilities for the editorial.
glucocorticoid use, the overall risk is much and definitive PJP. While there should Funding  This research received no specific grant from
lower (90% of the cases occurred at dose be no differential bias between the two any funding agency in the public, commercial or not-
levels above 15  mg/day), but these cases exposure groups that would affect one’s for-profit sectors.
illustrate that the risk of a certain dosage of ability to judge TMP/SMX effectiveness, Competing interests  KLW has been a consultant
glucocorticoids is likely modified by other the study might overestimate the risk of and/or investigator for Pfizer, AbbVie, Lilly, BMS,
PJP risk factors. The benefit of prophylaxis PJP and hence NNT calculations if some Galapagos and UCB. JWB has been a consultant for
Pfizer, Merck and R-Pharm.
is almost certainly different in an 80 year of these cases were only colonisation. It
old with lymphopaenia and vasculitis using is also unclear if these data can generalise Provenance and peer review  Commissioned;
externally peer reviewed.
10 mg/day of glucocorticoids as compared outside of Korea where the prevalence of
with a 50 year old with rheumatoid arthritis PJP colonisation (and therefore the risk of © Article author(s) (or their employer(s) unless
otherwise stated in the text of the article) 2018. All
(RA) using the same dose who lacks other developing disease) might differ. In addi- rights reserved. No commercial use is permitted unless
risk factors. We suggest that for patients tion, the analysis did not evaluate the risk otherwise expressly granted.
receiving prophylaxis that the treating of biological therapies. It is possible that
physician consider stopping TMP/SMX their use might modify the risk of gluco-
once doses have been tapered to 15 mg/day, corticoids, and that this could vary by
but that strong consideration be given to their mechanism of action. Further, the To cite Winthrop KL, Baddley JW. Ann Rheum Dis
Epub ahead of print: [please include Day Month Year].
continuing it until lower doses are achieved analysis only addressed risk among a high- doi:10.1136/annrheumdis-2017-212588
if other PJP risk factors are present. dose glucocorticoid using population, and
it did not report PJP incidence in rheuma-
tology patients who were not using high-
What is the risk-benefit of TMP/ dose glucocorticoids. The regimen studied
SMX? was single-strength daily TMP/Sulfa. The ►► http://​dx.​doi.​org/​10.​1136/​annrheumdis-​2017-​
The risk-benefit of TMP/SMX prophy- risk/benefit of intermittent (three times 211796
laxis has been debated, given the high a week) double strength TMP/SMX may Ann Rheum Dis 2018;0:1–3.
incidence of side effects reported with this be different, although experience from doi:10.1136/annrheumdis-2017-212588
compound.6 Prior analyses suggest that the other settings of immunosuppression
benefit outweighs the risk only in certain would suggest similar efficacy as daily References
1 Catherinot E, Lanternier F, Bougnoux ME, et al.
inflammatory disease conditions such single strength therapy.8 Last, it is unclear Pneumocystis jirovecii Pneumonia. Infect Dis Clin North
as vasculitis or dermatomyositis, largely if prophylaxis with other regimens (eg, Am 2010;24:107–38.
because the frequency of PJP is higher in dapsone, atovaquone) is of similar utility 2 Cettomai D, Gelber AC, Christopher-Stine L.
these conditions.4 7 Notably, prophylaxis or should be employed if patients cannot A survey of rheumatologists’ practice for
may not be favourable in rheumatoid tolerate TMP/SMX. prescribing pneumocystis prophylaxis. J Rheumatol
arthritis where the risk is low and number While these and other clinical ques- 3 Park JW, Curtis JR, Moon J, et al. Prophylactic effect
needed to treat (NNT) is much higher.4 tions remain, our opinion is that the of trimethoprim-sulfamethoxazole for pneumocystis
While this analysis is limited to those using current analysis provides strong guidance pneumonia in patients with rheumatic diseases exposed
high-dose glucocorticoids for greater in terms of who to select for prophy- to prolonged high-dose glucocorticoids. Ann Rheum
Dis 2017:10.1136/annrheumdis-2017-211796 [Epub
than 1 month, it supports these ideas and laxis. It supports the efficacy of TMP/
ahead of print 1 Nov 2017].
provides both NNT and numbers  needed SMX use among those starting regimens 4 Green H, Paul M, Vidal L, et al. Prophylaxis of
to harm (NNH) information within some of Prednisone>30 mg/day who continue Pneumocystis pneumonia in immunocompromised
disease subgroups. Overall, this cohort for greater than 1 month and particularly non-HIV-infected patients: systematic review and meta-

2 Winthrop KL, Baddley JW. Ann Rheum Dis Month 2018 Vol 0 No 0
Downloaded from on March 11, 2018 - Published by

analysis of randomized controlled trials. Mayo Clin Proc 6 Wolfe RM, Peacock JE. Pneumocystis Pneumonia and negative immunocompromised patients. Clin Microbiol
2007;82:1052–9. the Rheumatologist: Which Patients Are At Risk and Rev 2004;17:770–82.
5 Yale SH, Limper AH. Pneumocystis carinii pneumonia in How Can PCP Be Prevented? Curr Rheumatol Rep 8 Stern A, Green H, Paul M, et al. Prophylaxis
patients without acquired immunodeficiency syndrome: 2017;19:35. for Pneumocystis pneumonia (PCP) in non-HIV
associated illnesses and prior corticosteroid therapy. 7 Rodriguez M, Fishman JA. Prevention of infection due immunocompromised patients. Cochrane Database Syst
Mayo Clin Proc 1996;71:5–13. to Pneumocystis spp. in human immunodeficiency virus- Rev 2014:Cd005590.

Winthrop KL, Baddley JW. Ann Rheum Dis Month 2018 Vol 0 No 0 3
Downloaded from on March 11, 2018 - Published by

Pneumocystis and glucocorticoid use: to

prophylax or not to prophylax (and when?);
that is the question
Kevin L Winthrop and John W Baddley

Ann Rheum Dis published online February 19, 2018

Updated information and services can be found at:

These include:

References This article cites 6 articles, 2 of which you can access for free at:
Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.


To request permissions go to:

To order reprints go to:

To subscribe to BMJ go to: