Protective effect of different doses of trimethoprim-sulfamethoxazole prophylaxis for early severe infections among patients with antineutrophil ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Protective effect of different doses of
trimethoprim-sulfamethoxazole prophylaxis for early
severe infections among patients with antineutrophil
cytoplasmic autoantibody-associated vasculitis
D. Waki, K. Nishimura, T. Yoshida, N. Tanaka, K. Mizukawa,
M. Fukushima, O. Iri, M. Anegawa, H. Murabe, T. Yokota
Department of Endocrinology and ABSTRACT timal dose of TMP/SMX for preventing
Rheumatology, Kurashiki Central Objective. To analyse the protective ef- severe infections, especially consider-
Hospital, Okayama, Japan. fect of different doses of trimethoprim- ing renal impairment.
Daisuke Waki, MD sulfamethoxazole (TMP/SMX) prophy-
Keisuke Nishimura, MD, PhD laxis for early severe infections in anti- Introduction
Tomohiro Yoshida, MD
neutrophil cytoplasmic autoantibody- Disease control of antineutrophil cy-
Nozomi Tanaka, MD
Kaoru Mizukawa, MD associated vasculitis (AAV), consider- toplasm antibody (ANCA)-associated
Mayuko Fukushima, MD ing time-varying changes. vasculitis (AAV) and the prognosis of
Osamu Iri, MD Methods. In this retrospective obser- AAV have dramatically improved due
Motoko Anegawa, MD vational study, we assessed the protec- to the application of immunosuppres-
Hiroyuki Murabe, MD, PhD tive effect of TMP/SMX within the first sive therapy combined with corticos-
Toshihiko Yokota, MD, PhD 6 months of diagnosis among Japanese teroids and cyclophosphamide (CYC)
Please address correspondence to: patients with AAV. We included 250 con- or rituximab (RTX) (1-4). However,
Daisuke Waki, secutive patients with AAV who were controlling the side effects of treat-
Department of Endocrinology
and Rheumatology,
admitted to our hospital. The protective ment, particularly for severe life-
Kurashiki Central Hospital, effect of TMP/SMX against early severe threatening infections, remains chal-
1-1-1 Miwa, Kurashiki, infections was verified using Cox regres- lenging. Recent studies have shown
Okayama 710-8602, Japan. sion analysis along with potential con- that severe infection is a major cause
E-mail: dw15712@kchnet.or.jp founding factors. Cox regression with of death as well as active vasculitis.
Received on November 26, 2020; accepted inverse probability treatment weights The majority of severe infections oc-
in revised form on February 1, 2021. for early severe infections was also per- cur within the first 6 months of start-
Clin Exp Rheumatol 2021; 39 (Suppl. 129): formed as a sensitivity analysis. ing remission induction therapy (5-8).
S142-S148. Results. Cox regression analysis showed Although several risk factors for se-
© Copyright CLINICAL AND that the reduced TMP/SMX exposure vere infections have been suggested,
EXPERIMENTAL RHEUMATOLOGY 2021. group had a significant protective effect prophylactic methods against them
against early severe infections (stand- have not yet been established. Trimeth-
Key words: antineutrophil ard-dose group versus no TMP/SMX oprim-sulfamethoxazole (TMP/SMX)
cytoplasmic autoantibody, infection, group: hazard ratio [HR] 0.393, 95% is routinely prescribed as prophylaxis
trimethoprim-sulfamethoxazole, confidence interval [CI]: 0.139–1.11, for Pneumocystis jirovecii pneumonia
vasculitis p=0.077; reduced-dose group versus no in daily practice and is also expected
TMP/SMX group: HR 0.418, 95%CI: to reduce disease flare-ups in granulo-
0.216–0.807, p=0.009), even when con- matosis with polyangiitis (GPA) (9). A
sidering time-dependent changes. In the recent observational study has reported
sensitivity analysis, the reduced-dose that TMP/SMX reduced the risk of se-
group still had a significantly lower risk vere infections among European pa-
of early severe infections than the no tients with AAV with predominant GPA
TMP/SMX group (HR=0.393, 95%CI: (10). However, the protective effect of
0.177–0.873, p=0.022). During follow- TMP/SMX against severe infections in
up, 18.0% of the patients discontinued patients with AAV in different parts of
TMP/SMX due to side effects. the world remains unclear.
Conclusion. TMP/SMX is highly ef- Although the side effects of TMP/SMX
fective in preventing severe infections are relatively frequent, previous stud-
among patients with AAV despite the ies did not consider TMP/SMX expo-
high incidence of side effects. Further sure as a time-varying factor; some
Competing interests: none declared. studies are needed to determine the op- studies have reported that as many as
S-142 Clinical and Experimental Rheumatology 2021Vasculitis: prevention of infections / D. Waki et al.
20% of patients needed to discontinue
TMP/SMX because of side effects
(9, 11). Moreover, we frequently pre-
scribe TMP/SMX at a reduced dosage
to effectively prevent Pneumocystis
jirovecii pneumonia due to concerns
about the high incidence of side effects.
This study aimed to assess the protec-
tive effect of TMP/SMX for severe in-
fections in Japanese patients with AAV
considering time-varying changes and
to investigate the differences in the
protective effect among different TMP/
SMX dosage groups.
Materials and methods
Study population
This study was conducted as a ret-
rospective observational study. We
Fig. 1. Flow diagram of patient selection.
screened newly diagnosed patients ad- A total of 254 patients were included in this retrospective observational study after excluding 20 pa-
mitted to our hospital for AAV between tients due to various reasons. The patients were categorised into three groups according to the received
January 2000 and January 2020 and dosage of trimethoprim-sulfamethoxazole (TMP/SMX) at 1 month after diagnosis.
collected the data retrospectively. All
patients were diagnosed according to an opt-out methodology. Patient data tis jirovecii pneumonia. The prescriber
the European Medicines Agency algo- were recorded at each follow-up visit reduced the dose by half for patients
rithm as having GPA, microscopic pol- and collected from medical records un- with impaired renal function (creati-
yangiitis (MPA), eosinophilic granulo- til loss to follow-up or 6 months after nine clearance ≤30 ml/min) according
matosis with polyangiitis (EGPA) (12). diagnosis. The date of diagnosis was to drug package inserts of pharmaceu-
Clinical manifestations and laboratory defined as the day that remission induc- ticals and medical devices agency in
data for comparing patient character- tion therapy was started. Japan. Patients with severely impaired
istics were recorded at the time of di- All included patients had not received renal function (creatinine clearance ≤15
agnosis except for the following: hav- any immunosuppressive therapy be- ml/min) were generally not prescribed
ing steroid-induced diabetes mellites fore the diagnosis of AAV. Remission TMP/SMX according to drug pack-
(DM), received haemodialysis (HD), induction therapy included oral or in- age inserts. The decision to reduce the
cumulative doses of corticosteroids, travenous CYC, RTX, mycophenolate dosage of TMP/SMX or change the
vasculitis damage index, and receiving mofetil, methotrexate, azathioprine, prophylactic drug was the physician’s
TMP/SMX dose. calcineurin inhibitors, intravenous im- discretion when side effects were ob-
Patients with incomplete or missing munoglobulin, and plasma exchange, in served. We analysed the sample’s risk
records regarding vasculitis activity or addition to corticosteroids. Treatment factor for severe infections within 6
treatment regimen were excluded from regimens were selected at the discretion months, including TMP/SMX expo-
final analyses. Patients with missing of each physician. In general, initial sure. Because TMP/SMX exposure is a
records regarding the history of infec- doses of corticosteroids were contin- time-dependent variable due to the high
tious episodes for more than 6 months ued for 2 weeks, with gradual tapering incidence of adverse effects, which
until loss to follow up were excluded. of the dose every 2 weeks. Prophy- mainly occur within the first month
Patients with classic polyarteritis no- lactic methods against Pneumocystis of prescription, patient characteristics
dosa (cPAN), unclassified vasculitis, jirovecii pneumonia were also used at were compared among different TMP/
or secondary vasculitis who did not the discretion of each physician. Since SMX exposure groups at 1 month after
receive immunosuppressive treatment, there is no authorised prophylaxis regi- diagnosis.
or who died within 1 week after diag- men of TMP/SMX against Pneumocys-
nosis, were also excluded. tis jirovecii pneumonia regarding renal Definition
Finally, 250 patients were included in impairment, physicians generally chose Among the different TMP/SMX dos-
this study. Our research was performed from two regimens: 160 mg/800 mg age groups, the standard-dose group
according to the principles outlined in TMP/SMX three times a week or 80 was defined as receiving 160 mg/800
the Declaration of Helsinki and ap- mg/400 mg TMP/SMX once daily for mg TMP/SMX three times a week or
proved by the ethics committee of our patients without renal impairment (13, 80 mg/400 mg TMP/SMX once daily.
hospital (approval number: 3430). In- 14). These are the most common dosag- The reduced-dose group received a
formed consent was obtained through es for prophylaxis against Pneumocys- lower dosage of TMP/SMX per week,
Clinical and Experimental Rheumatology 2021 S-143Vasculitis: prevention of infections / D. Waki et al.
Table I. Patient characteristics.
Standard-dose group Reduced-dose group No TMP/SMX Reduced- TMP/SMX
(n=40) (n=167) group (n=43) dose group exposure
vs. vs.
No TMP/ No TMP/
SMX group SMX group
p-value p-value
Sex 0.734 0.504
Female, n (%) 26 (65.0) 91 (54.5) 22 (51.2)
Male, n (%) 14 (35.0) 76 (45.5) 21 (48.8)
Age at diagnosis, median [IQR] 66.5 [60.3-76.0] 76.0 [68.0-81.0] 75.0 [67.0-81.5] 0.743 0.233
Past history of smoking, n (%) 15 (40.5) 70 (46.1) 19 (46.3) 1.000 0.864
Past history of CKD, n (%) 3 (7.9) 35 (22.0) 7 (17.1) 0.667 0.666
Past history of lung disease, n (%) 12 (30.8) 38 (23.0) 11 (25.6) 0.692 1.000
Steroid DM or past history of DM, n (%) 15 (38.5) 95 (57.2) 17 (39.5) 0.041 0.07
ANCA, n (%) 1.000 0.862
MPO-ANCA/p-ANCA 28 (70.0) 137 (82.0) 36 (83.7)
PR3-ANCA/c-ANCA 5 (12.5) 14 (8.4) 3 (7.0)
Both negative 7 (17.5) 16 (9.6) 4 (9.3)
Type of AAV, n (%) 0.567 0.323
MPA 20 (50.0) 118 (70.7) 31 (72.1)
GPA 14 (35.0) 29 (17.4) 5 (11.6)
EGPA 6 (15.0) 20 (12.0) 7 (16.3)
BVAS†,
median [IQR] 8.0 [6.0-16.0] 14.0 [12.0-18.0] 15.0 [12.0-18.0] 0.538 0.197
VDI, median [IQR] 2.0 [1.0-3.0] 3.0 [2.0-3.0] 3.0 [2.0-3.0] 0.901 0.364
Serum IgG (mg/dL), median [IQR] 1876.0 [1496.5-2161.0] 1537.0 [1274.5-2023.0] 1531.0 [1217.5-1891.0] 0.549 0.48
Serum lymphocyte (/μL), median [IQR] 1413.3 [1135.1-1701.5] 1155.6 [869.2-1593.2] 1089.0 [780.5-1378.7] 0.169 0.074
CRP (mg/dL), median [IQR] 7.9 [4.8-11.4] 6.4 [1.1-10.7] 6.6 [2.3-10.9] 0.427 0.759
Serum creatinine (mg/dL), median [IQR] 0.66 [0.54-0.74] 1.56 [0.67-3.76] 1.50 [0.65-2.95] 0.628 0.302
eGFR (mL/min per 1.73 m2) median [IQR] 68.2 [58.4-85.9] 23.9 [8.5-65.5] 24.8 [11.6-66.6] 0.597 0.314
Revised FFS‡ ≥2, n (%) 14 (35.0) 128 (76.6) 30 (69.8) 0.428 0.859
Renal involvement, n (%) 8 (20.0) 115 (68.9) 31 (72.1) 0.853 0.124
Pulmonary involvement, n (%) 28 (70.0) 68 (40.7) 14 (32.6) 0.383 0.181
Initial doses of CS/weight (mg/kg/day)§, median [IQR] 0.99 [0.80-1.00] 0.75 [0.63-1.00] 0.95 [0.66-1.00] 0.023 0.07
MPSL pulse therapy, n (%) 10 (25.0) 104 (62.3) 25 ( 58.1) 0.726 1.000
Cumulative CS/weight at 1 month (mg/kg)§, 25.3 [21.1-27.3] 20.6 [17.2-26.7] 25.1 [17.7-27.7] 0.08 0.237
median [IQR]
CYC 21 (52.5) 37 (22.2) 9 (20.9) 1.000 0.351
RTX 0 9 (5.4) 7 (16.3) 0.025 0.002
Other immunosuppressants¶ 4 (10.0) 16 (9.6) 4 (9.3) 1.000 0.777
Received HD, n (%) 0 13 (7.8) 3 (7.0) 1.000 1.000
Interquartile range (IQR): data are presented as median [1st quartile to 3rd quartile].
†
BVAS = Birmingham Vasculitis Activity Score version 3 (16).
‡
revised FFS = the revised Five-Factor Scores reported in 2009 (17).
§
Different types of corticosteroids were converted to equivalent doses of prednisolone.
¶
mycophenolate mofetil, methotrexate, azathioprine, and calcineurin inhibitors.
TMP/SMX: trimethoprim-sulfamethoxazole; SD: standard deviation; CKD: chronic kidney disease; DM: diabetes mellitus; ANCA: anti-neutrophil cy-
toplasmic antibody; MPO: myeloperoxidase; PR3: perinuclear proteinase-3; AAV: antineutrophil cytoplasmic autoantibody-associated vasculitis; MPA:
microscopic polyangiitis; GPA: granulomatosis with polyangiitis; EGPA: eosinophilic granulomatosis with polyangiitis; cPAN: classic polyarteritis nodosa;
BVAS: Birmingham Vasculitis Activity Score; VDI: vasculitis damage index; IgG: Immunoglobulin G; CRP: C-reactive protein; IQR: interquartile range;
eGFR: estimated glomerular filtration rate; FFS: Five Factor Score; CS: corticosteroids; MPSL: methylprednisolone; CYC: cyclophosphamide; RTX: ritux-
imab; HD: haemodialysis.
and the no TMP/SMX group did not eGFR equation based on standardised cording to BVAS version 3, and pulmo-
receive TMP/SMX. Patients with any serum creatinine levels (15). Past his- nary and renal involvement were also
TMP/SMX exposure were assigned to tory of DM was defined as receiving scored according to the BVAS (16).
the combined standard-dose group and antidiabetic treatment at disease onset, The Five-Factor Score (revised FFS)
reduced-dose group. and steroid DM was defined as receiv- at diagnosis was calculated according
A history of chronic kidney disease ing antidiabetic treatment within 1 to Five-Factor Score 2009 (17). Indi-
(CKD) was defined as an estimated glo- month after diagnosis. Received HD rect immunofluorescence or antigen-
merular filtration rate (eGFR)Vasculitis: prevention of infections / D. Waki et al. hospitalisation or intravenous antibiot- sure groups and the variables that were sis showed a tendency to be higher in ics. Disseminated herpes zoster recur- significantly different (p.value
Vasculitis: prevention of infections / D. Waki et al.
Table II. Multivariate analysis for early severe infections. these patients had received a reduced
dose of TMP/SMX because of severe
Risk factor Hazard ratio 95% CI p-value
renal insufficiency. Other severe in-
TMP/SMX exposurea fections included sepsis of unknown
Standard-dose group vs. No TMP/SMX group 0.393 (0.139-1.11) 0.077 origin (n=6), cytomegalovirus infec-
Reduced-dose group vs. No TMP/SMX group 0.418 (0.216-0.807) 0.009 tions (n=5), skin and soft tissue infec-
Steroid DM or past history of DM 0.902 (0.493-1.649) 0.737
Serum lymphocyte countb 0.943 (0.890-0.998) 0.043 tions (n=4), disseminated herpes zoster
Initial dose of CS/weightc 2.446 (0.536-11.580) 0.248 recurrences (n=3), fungal infections
Receiving RTX 1.923 (0.804-4.597) 0.142 (n=2), biliary tract infections (n=2),
a
intestinal infections (n=2), mycobacte-
TMP/SMX exposure was entered as a discrete time-varying covariate.
b
For every 100/µL increase. rium avium complex infections (n=1),
c
Different types of corticosteroids were converted to equivalent doses of prednisolone. disseminated nocardia infections (n=1),
TMP/SMX: trimethoprim-sulfamethoxazole; DM: diabetes mellitus; CS: corticosteroids; CI: confidence urinary tract infections (n=1), and sup-
interval; RTX: rituximab.
purative parotitis (n=1).
The exact pathogen was identified in 31
Fig. 2. Severe infection-free instances of severe infections. P. aer-
survival rates within 6 months
among the three groups.
uginosa and cytomegalovirus were the
A: Kaplan-Meier curve show- most common pathogens (both n=5),
ing the reduced-dose group followed by E. coli (n=4), herpes zos-
had a significantly lower risk ter (n=3), S. aureus (n=3), P. jirovecii
of early severe infections
than the no trimethoprim-sul-
(n=2), Aspergillus spp. (n=2), S. pneu-
famethoxazole (TMP/SMX) moniae (n=1), Nocardia cyriacigeor-
group (Bonferroni-corrected gica (n=1), Mycobacterium avium com-
p=0.012). plex (n=1), K. oxytoca (n=1), E. faeci-
B: The adjusted Kaplan-Meier
curve using IPTW among the um + P. aeruginosa (n=1), E. cloacae
three groups. (n=1), and C. perfringens (n=1).
Reasons for discontinuing TMP/SMX
Among the total sample, 45 patients
(18.0%) discontinued TMP/SMX
within 6 months. The most frequent
reason for discontinuation was cyto-
penia (n=19), followed by rash (n=9),
liver dysfunction (n=7), hyperkalaemia
(n=4), unknown (n=4), renal dysfunc-
tion (n=1), and the end of treatment for
AAV at the patient’s request (n=1). Af-
ter discontinuation of TMP/SMX, all
patients received alternative prophy-
laxis for Pneumocystis jirovecii pneu-
monia.
Discussion
nificantly lower risk of early severe in- group could not be confirmed statisti- Our study demonstrated that reduced
fections than the no TMP/SMX group cally (HR 0.595, 95%CI: 0.196–1.806, TMP/SMX exposure has a protective
(Bonferroni-corrected p=0.012) (Fig. p=0.359). The adjusted Kaplan-Meier effect for severe infections within the
2A). curves using IPTW are shown in Fig. first 6 months in Japanese patients with
2B. AAV, considering time-varying chang-
Sensitivity analysis es. These results are in line with those
The adjusted Cox regression analysis Aetiologies of early severe infections of previous studies.
using IPTW revealed that TMP/SMX We identified 45 severe infections dur- Despite remarkable advances in treat-
still had a significant protective effect ing 6 months after diagnosis (Table IV). ment, the mortality rate for patients
for early severe infections in the re- Bacterial pneumonia was most frequent with AAV remains higher than that for
duced-dose group (HR 0.393, 95%CI: type of infection in the total patient age- and sex-matched persons in the
0.177–0.873, p=0.022) (Table III). Pro- cohort (n=15, 33.3%). Unfortunately, general population and the main cause
tective effect of TMP/SMX for early two patients in the TMP/SMX group of death within the first year is infec-
severe infections in the standard-dose had Pneumocystis jirovecii pneumonia; tion (27). The current European League
S-146 Clinical and Experimental Rheumatology 2021Vasculitis: prevention of infections / D. Waki et al.
Table III. Multivariate analysis for early severe infections using an inverse probability- previous reports included relapse pa-
weighted Cox regression model. tients, this study included only new-
Risk factor Hazard ratio 95% CI p-value
onset patients with AAV; the results
were therefore not affected by previous
TMP/SMX exposure† treatment.
Standard-dose group vs. No TMP/SMX group 0.595 (0.196-1.806) 0.359 Our study could not prove the presence
Reduced-dose group vs. No TMP/SMX group 0.393 (0.177-0.873) 0.022
Steroid DM or past history of DM 1.001 (0.441-2.276) 0.998
of a significant protective effect against
Serum lymphocyte count‡ 0.917 (0.845-0.995) 0.037 severe early infection in the standard-
Initial dose of CS/weight§ 3.636 (0.701-18.666) 0.122 dose group, although the survival analy-
Receiving RTX 1.588 (0.590-4.279) 0.360 sis implied that the standard-dose group
†
seemed to have the same protective ef-
TMP/SMX exposure was entered as a discrete time-varying covariate.
‡
For every 100/µL increase. fect as the reduced-dose group. We be-
§
Different types of corticosteroids were converted to equivalent doses of prednisolone. lieve that the reason for these results
TMP/SMX: trimethoprim-sulfamethoxazole; DM: diabetes mellitus; CS: corticosteroids; CI: confidence was derived from the relatively small
interval.
sample size in the standard-dose group.
One reason for the inadequate sample
Table IV. Summary of early severe infections.
size was that in most cases receiving a
Infections Total Standard-dose Reduced-dose No TMP/ standard dose of TMP/SMX, the dos-
n (%) group group SMX group age had to be reduced or discontinued
(n=5) (n=25) (n=15) due to side effects during the course of
Bacterial pneumonia 15 (33.3) 3 (60.0) 9 (36.0) 3 (20.0)
study (34.7% among the standard-dose
Sepsis (unknown origin) 6 (13.3) 1 (20.0) 2 (8.0) 3 (20.0) group during the study period, data not
Cytomegalovirus infections 5 (11.1) 0 2 (8.0) 3 (20.0) shown).
Skin and soft tissue infections 4 (8.9) 0 2 (8.0) 2 (13.3) We found that 18.0% of patients needed
Disseminated herpes zoster infections 3 (6.7) 0 1 (4.0) 2 (13.3)
to discontinue TMP/SMX because of
Fungal infections 2 (4.4) 1 (20.0) 1 (4.0) 0
Biliary tract infections 2 (4.4) 0 2 (8.0) 0 side effects. Although not all side ef-
Pneumocystis jirovecii pneumonia 2 (4.4) 0 2 (8.0) 0 fects may have been due to TMP/SMX
Intestinal infections 2 (4.4) 0 1 (4.0) 1 (6.7) and most of the patients received a re-
Mycobacterium avium complex infections 1 (2.2) 0 0 1 (6.7) duced dose, the frequency of discon-
Disseminated nocardia infections 1 (2.2) 0 1 (4.0) 0
Urinary tract infections 1 (2.2) 0 1 (4.0) 0 tinuation was generally consistent with
Suppurative parotitis 1 (2.2) 0 1 (4.0) 0 that of previous studies (9, 30-32). The
results of some studies have indicated
TMP/SMX: trimethoprim-sulfamethoxazole. that the side effects of TMP/SMX were
reduced by a reduced-dose regimen;
Against Rheumatism/European Renal RTX (10). Previous studies have fo- however, our sample had a high num-
Association - European Dialysis and cused primarily on European patients ber of patients with decreased kidney
Transplant Association recommenda- with AAV, in whom GPA is predomi- function, which may have increased the
tions encourage the use of TMP/SMX nant; these studies also did not consider toxicity of TMP/SMX (11, 32).
as prophylaxis against infection with time-varying changes related to TMP/ To the best of our knowledge, there
Pneumocystis jirovecii in all patients SMX exposure, although discontinu- is no authorised prophylaxis regimen
with AAV who are being treated with ation is relatively frequent due to side of TMP/SMX against Pneumocystis
cyclophosphamide (28). However, lit- effects. Moreover, most life-threatening jirovecii pneumonia for rheumatic dis-
tle is known about its protecting effect severe infections occur within the first 6 eases patients regarding renal impair-
against severe infections. months of starting remission induction ment. Although one open randomised
In 1996, a double-blind, placebo- therapy (5-8, 20). Our study provides trial reported the equal efficacy for pre-
controlled trial assessed the efficacy robust evidence of the protective ef- venting Pneumocystis jirovecii infection
of TMP/SMX in preventing relapses fect of TMP/SMX against early severe between the standard TMP/SMX dose
among patients with GPA and found infections among patients with AAV in group and half-dose group at week 24 in
that the number of infectious episodes an MPA-predominant sample and con- patients with various systemic rheumatic
per patient within 24 months was sig- sidering the time-varying changes that diseases, there were two cases of Pneu-
nificantly lower in the TMP/SMX occur with TMP/SMX exposure. We mocystis jirovecii infection in the re-
group (9). Kronbichler et al. reported believe our study revealed useful infor- duced-dose group in our study (32). The
that the use of TMP/SMX was associ- mation in the context of the increasing optimal TMP/SMX dose for the preven-
ated with a lower frequency of severe incidence of MPA worldwide (29). In tion of infections, including Pneumocys-
infections during follow-up (mean time addition, we found that a reduced dose tis jirovecii, warrants further study.
of 22.7 months) in new-onset and re- of TMP/SMX was effective in the pre- We acknowledge that there were several
lapsed patients with AAV receiving vention of severe infections. Whereas limitations to our study. First, given its
Clinical and Experimental Rheumatology 2021 S-147Vasculitis: prevention of infections / D. Waki et al.
retrospective single-centre observation- M: Predictors for mortality in patients with Clin J Am Soc Nephrol 2013; 8: 416-23.
al design, there could have been both antineutrophil cytoplasmic autoantibody- 20. MOHAMMAD AJ, SEGELMARK M, SMITH
associated vasculitis: a study of 398 Chinese R et al.: Severe infection in antineutrophil
selection and recall bias. Second, the patients. J Rheumatol 2014; 41: 1849-55. cytoplasmic antibody-associated vasculitis.
study may have been unable to compre- 7. LITTLE MA, NIGHTINGALE P, VERBURGH J Rheumatol 2017; 44: 1468-75.
hensively detect potential confounding CA et al.: Early mortality in systemic vascu- 21. MCGREGOR JG, NEGRETE-LOPEZ R, POUL-
litis: Relative contribution of adverse events TON CJ et al.: Adverse events and infectious
factors among all groups, considering and active vasculitis. Ann Rheum Dis 2010; burden, microbes and temporal outline from
TMP/SMX exposure as a time-varying 69: 1036-43. immunosuppressive therapy in antineutro-
factor. However, it seems reasonable to 8. CHARLIER C, HENEGAR C, LAUNAY O et al.: phil cytoplasmic antibody-associated vascu-
compare the background among groups Risk factors for major infections in Wegener litis with native renal function. Nephrol Dial
granulomatosis: analysis of 113 patients. Ann Transplant 2015; 30 Suppl 1: i171-81.
at 1 month in the present study in the Rheum Dis 2009; 68: 658-63. 22. CHEN M, YU F, ZHANG Y, ZHAO MH: Anti-
context of previous reports. Third, the 9. STEGEMAN CA, TERVAERT JW, JONG PE DE, neutrophil cytoplasmic autoantibody-asso-
prescription patterns of TMP/SMX var- KALLENBERG CG: Trimethoprim-sulfameth- ciated vasculitis in older patients. Medicine
ied in our cohort. Prospective compara- oxazole (co-trimoxazole) for the prevention (Baltimore) 2008; 87: 203-9.
of relapses of Wegener’s granulomatosis. 23. YOO J, JUNG SM, SONG JJ, PARK Y, LEE S:
tive studies with defined TMP/SMX Dutch Co-Trimoxazole Wegener Study Birmingham vasculitis activity and chest
dosages according to renal function are Group. N Engl J Med 1996; 335: 16-20. manifestation at diagnosis can predict hospi-
warranted in the future. 10. KRONBICHLER A, KERSCHBAUM J, GOPALU- talised infection in ANCA-associated vascu-
In conclusion, TMP/SMX is highly ef- NI S et al.: Trimethoprim-sulfamethoxazole litis. Clin Rheumatol 2018; 37: 2133-41.
prophylaxis prevents severe/life-threatening 24. YANG L, XIE H, LIU Z et al.: Risk factors for
fective in preventing severe infections infections following rituximab in antineutro- infectious complications of ANCA-associat-
as well as Pneumocystis jirovecii infec- phil cytoplasm antibody-associated vasculi- ed vasculitis: a cohort study. BMC Nephrol
tion in patients with AAV, despite the tis. Ann Rheum Dis 2018; 77: 1440-47. 2018; 19: 138.
high incidence of side effects. Further 11. TAKENAKA K, KOMIYA Y, OTA M, YAMAZA- 25. WATANABE-IMAI K, HARIGAI M, SADA K
KI H, NAGASAKA K: A dose-escalation regi- et al.: Clinical characteristics of and risk
prospective studies will be warranted men of trimethoprim–sulfamethoxazole is factors for serious infection in Japanese pa-
to determine the optimal dose of TMP/ tolerable for prophylaxis against Pneumo- tients within six months of remission induc-
SMX for the prevention of severe infec- cystis jiroveci pneumonia in rheumatic dis- tion therapy for antineutrophil cytoplasmic
eases. Mod Rheumatol 2013; 23: 752-58. antibody-associated vasculitis registered in
tions, especially regarding renal impair- 12. WATTS R, LANE S, HANSLIK T et al.: Devel- a nationwide, prospective, inception cohort.
ment. opment and validation of a consensus meth- Mod Rheumatol 2017; 27: 646-51.
odology for the classification of the ANCA- 26. LAFARGE A, JOSEPH A, PAGNOUX C et al.:
Acknowledgments associated vasculitides and polyarteritis no- Predictive factors of severe infections in pa-
dosa for epidemiological studies. Ann Rheum tients with systemic necrotizing vasculitides:
The authors wish to thank Dr Akira On- Dis 2007; 66: 222-27. data from 733 patients enrolled in five rand-
ishi, Department of Rheumatology and 13. LIMPER AH, KNOX KS, SAROSI GA et al.: omized controlled trials of the French Vascu-
Clinical Immunology, Kobe University An Official American Thoracic Society State- litis Study Group. Rheumatology 2020; 59:
Graduate School of Medicine, and Dr ment: Treatment of fungal infections in adult 2250-7.
pulmonary and critical care patients. Am J 27. FLOSSMANN O, BERDEN A, GROOT K DE et
Hironobu Tokumasu, Department of Respir Crit Care Med 2011; 183: 96-128. al.: Long-term patient survival in ANCA-
Management, Clinical Research Cent- 14. HUGHES WT, RIVERA GK, SCHELL MJ, associated vasculitis. Ann Rheum Dis 2011;
er, Kurashiki Central Hospital, for their THORNTON D, LOTT L: Successful intermit- 70: 488-94.
help in the data analysis. tent chemoprophylaxis for pneumocystis 28. YATES M, WATTS RA, BAJEMA IM et al.:
carinii pneumonitis. N Engl J Med 1987; 316: EULAR/ERA-EDTA recommendations for
1627-32. the management of ANCA-associated vascu-
References 15. MATSUO S, IMAI E, HORIO M et al.: Revised litis. Ann Rheum Dis 2016; 75: 1583-94.
1. JAYNE D, RASMUSSEN N: Twenty-five years equations for estimated GFR from serum cre- 29. FELICETTI M, TREPPO E, POSARELLI C et al.:
of European Union collaboration in ANCA- atinine in Japan. Am J Kidney Dis 2009; 53: One year in review 2020: vasculitis. Clin Exp
associated vasculitis research. Nephrol Dial 982-92. Rheumatol 2020; 38 (Suppl. 124): S3-14.
Transplant 2015; 30: i1-i7. 16. MUKHTYAR C, LEE R, BROWN D et al.: 30. BRAKEMEIER S, PFAU A, ZUKUNFT B, BUD-
2. STONE JH, MERKEL PA, SPIERA R et al.: Modification and validation of the Birming- DE K, NICKEL P: Prophylaxis and treatment
Rituximab versus cyclophosphamide for ham vasculitis activity score (version 3). Ann of Pneumocystis Jirovecii pneumonia after
ANCA-associated vasculitis. N Engl J Med Rheum Dis 2009; 68: 1827-32. solid organ transplantation. Pharmacol Res
2010; 363: 221-32. 17. GUILLEVIN L, PAGNOUX C, SEROR R, MAHR 2018; 134: 61-7.
3. GROOT K DE, HARPER L, JAYNE DRW et al.: A, MOUTHON L, TOUMELIN P LE: The five- 31. PARK JW, CURTIS JR, MOON J, SONG YW, KIM
Pulse Versus Daily Oral Cyclophosphamide factor score revisited: Assessment of prog- S, LEE EB: Prophylactic effect of trimeth-
for Induction of Remission in Antineutrophil noses of systemic necrotizing vasculitides oprim-sulfamethoxazole for pneumocystis
Cytoplasmic Antibody-Associated Vasculi- based on the french vasculitis study group pneumonia in patients with rheumatic dis-
tis. Ann Intern Med 2009; 150: 670-80. (FVSG) cohort. Medicine (Baltimore) 2011; eases exposed to prolonged high-dose gluco-
4. MONTI S, BOND M, FELICETTI M et al.: 90: 19-27. corticoids. Ann Rheum Dis 2018; 77: 644-9.
One year in review 2019: vasculitis. Clin Exp 18. XIE J, LIU C: Adjusted Kaplan-Meier estima- 32. UTSUNOMIYA M, DOBASHI H, ODANI T et
Rheumatol 2019; 37 (Suppl. 117): S3-19. tor and log-rank test with inverse probability al.: Optimal regimens of sulfamethoxazole-
5. KRONBICHLER A, JAYNE DRW, MAYER G: of treatment weighting for survival data. Stat trimethoprim for chemoprophylaxis of
Frequency, risk factors and prophylaxis of Med 2005; 24: 3089-110. Pneumocystis pneumonia in patients with
infection in ANCA-associated vasculitis. Eur 19. GOUPIL R, BRACHEMI S, NADEAU ACF et al.: systemic rheumatic diseases: Results from a
J Clin Invest 2015; 45: 346-68. Lymphopenia and treatment-related infectious non-blinded, randomized controlled trial. Ar-
6. LAI Q, MA T, LI Z, CHANG D, ZHAO M, CHEN complications in ANCA-associated vasculitis. thritis Res Ther 2017; 19: 1-10.
S-148 Clinical and Experimental Rheumatology 2021You can also read
