Download Risk Factors for Isolation of Streptococcus pneumoniae with

437
Risk Factors for Isolation of Streptococcus pneumoniae with Decreased
Susceptibility to Penicillin G from Patients Infected with Human
Immunodeficiency Virus
J. L. Meynard, F. Barbut, L. Blum, M. Guiguet,
C. Chouaid, M. C. Meyohas, o. Picard, J. C. Petit,
and J. Frottier
From the Service des Maladies Infectieuses et Tropicales, Service de
Microbiologie, Service de Medecine Interne, and Service de
Pneumologie, H6pital Saint-Antoine; Institut National de la Sante et
Recherche Medical Unite 263, Paris, France
We conducted a retrospective study of all hospitalized human immunodeficiency virus (HIV)infected patients from whom a strain of Streptococcus pneumoniae was isolated (n = 45) between
January 1992 and September 1994, in order to determine the clinical manifestations and outcome
of and risk factors for infection by S. pneumoniae with decreased susceptibility to penicillin G.
Such strains were isolated from 14 patients (31%), of whom 8 had pneumonia, 2 had bronchial
superinfection, 2 had sinusitis, and 2 were colonized. All infected patients made a clinical recovery
regardless of the MIC of the isolate. Indexes of HIV disease stage (CD4+ cell count and p24
antigenemia), antiretroviral treatment, and hospital admission in the previous 3 months did not
influence the susceptibility of the isolates. For HIV-infected patients, treatment with antibacterial
agents-particularly trimethoprim-sulfamethoxazole-in the previous 3 months is associated with
an increased risk for isolation of S. pneumoniae with decreased susceptibility to penicillin G (relative
risk, 5.0; 95% confidence interval, 1.9-13.3).
Infection by HIV has various effects on physiological mechanisms of cellular and humoral immunity and predisposes patients to bacterial infections [1]. Several studies have shown
that Streptococcus pneumoniae is the main cause of lower
respiratory tract infections in HIV ~infected adults [2]. For example, the incidence of pneumococcal pneumonia is 5.5 to
17.5 times higher in patients with AIDS than in the general
population in Boston and New York [3, 4]. Clinical manifestations are similar to those in the general population of patients
with pneumococcal pneumonia, although two features, the recurrent and the bacteremic nature of the infection, have been
highlighted by certain authors [2-4]. There have been reports
of increased mortality associated with bacteremic forms of S.
pneumoniae infection in patients with AIDS [5].
Resistance to penicillin has increased greatly in the last few
years among clinical S. pneumoniae isolates in many countries.
In France, the estimated incidence of S. pneumoniae with decreased susceptibility to penicillin (PRP; herein the abbreviation refers to both penicillin-resistant strains and those with
decreased susceptibility to the drug) was 17% in 1990 and
25.1 % in 1994 [6].
An epidemiological survey designed to identify risk factors
for acquiring PRP has shown that HIV infection multiplies the
risk of invasive infection by a factor of > 3 [7]. The aims of
Received 6 July 1995; revised 31 October 1995.
Reprints or correspondence: Dr. Jean-Luc Meynard, Service des Maladies
Infectieuses et Tropica1es, H6pita1 Saint-Antoine, 184 rue du Faubourg SaintAntoine, 75012 Paris, France.
Clinical Infectious Diseases 1996;22:437-40
© 1996 by The University of Chicago. All rights reserved.
1058--4838/96/2203 -0006$02.00
this study were (1) to determine the clinical manifestations,
treatment, and outcome of infections due to PRP in HIV-infected patients and (2) to identify these patients' risk factors
for acquiring PRP.
Patients and Methods
Patients
Charts of all HIV-infected patients admitted to three units
in the Saint-Antoine Hospital between January 1992 and September 1994 and from whom S. pneumoniae was isolated
(n = 45) were reviewed. The following variables were recorded: age, sex, risk factor for HIV infection, smoking status,
CD4+ and CD8+ cell counts, neutrophil counts, p24 antigen
status, pentamidine inhalation, antibiotic therapy and admission
to a hospital during the previous 3 months, antiretroviral treatment, the sites from which the organism was isolated, clinical
manifestations, recurrence of the infection, treatment, and outcome. The CD4+ and CD8+ cell counts cited are those measured within 3 months of isolation of the bacterium or during
the infectious episode. Neutrophil counts are those measured
within 24 hours of isolation of the bacterium.
Clinical manifestations were classified as sinusitis (radiological diagnosis), bronchial superinfection (cough, expectoration,
fever, and no evocative radiological images), pneumonia (fever
and evidence of a local or diffuse infiltrate on chest roentgenography), and colonization (isolation of organisms from respiratory tract specimens; no radiological abnormalities or fever).
Methods
Antimicrobial susceptibility testing was done with an agar
disk diffusion method (Mueller-Hinton agar containing 5%
438
cm
Meynard et al.
1996; 22 (March)
Table 1. Clinical and laboratory findings for 14 patients infected or colonized by S. pneumoniae with decreased susceptibility to penicillin
G (group 1).
Patient
no.
First isolation (F)
or relapse (R)
No.ofCD4
cells per mm3
Isolation
site
Diagnosis
I
2
3
4
5
3
3
25
294
2
BA
BA, SP
BC
BA
BA
PN
PN
PN
PN
PN
F
R
F
F
F
6
7
8
9
127
92
BA
BC, SP
BA
SP
BA
SP
BA
SP
BC
SI bronch
PN
PN
Sinusitis
SI bronch
Col
Col
Sinusitis
PN
F
R
F
F
F
R
F
F
F
10
II
12
13
14
11
18
3
25
40
10
150
Treatment
Amox (6 g/24 h)
Teico (6 mg/[kg' 24 h])
Amox (6 g/24 h)
Amox (3 g/24 h)
Ctax (3 g124 h), Teico
(6 mg/[kg' 24 h])
Amox (6 g124 h)
Ctax (4 g/24 h)
Prist (3 g/24 h)
Cfix
Amox (2 g/24 h)
None
None
Prist (3 g/24 h)
Amox (6 g124 h)
MIC
(j-tg/mL)
2
2
1.5
1.5
2
1.5
2
1.5
0.1
0.5
0.5
0.25
0.25
0.2
Serotype
ND
23 F
23
14
6
23 F
23 F
23 F
23 F
ND
ND
ND
23
23
NOTE. The outcome was favorable for each patient. Amox = amoxicillin; BA = bronchial aspirate; BC = blood (culture); Cfix = cefixime; Col =
colonization; Ctax = cefotaxime; ND = not determined; PN = pneumonia; Prist = pristinamycin; SI bronch = bronchial superinfection; SP = sputum specimen;
Teico = teicoplanin.
horse blood, incubated for 18 hours at 37°C in air containing
5% CO2 ), according to the recommendations of the Comite de
l'Antibiogramme de la Societe Fran<;aise de Microbiologie [8].
Resistance of S. pneumoniae to penicillin G was suspected if
the diameter of the zone of inhibition around the 5-l1g oxacillin
disk was :::;25 mm. Resistance was confirmed by determination
of MIC by the Etest method (AB BIODISK, Solna, Sweden)
[9]. Isolates were designated as susceptible (MIC, <0.06 I1g/
mL), intermediately resistant (MIC, 0.12-1 I1g/mL), or highly
resistant to penicillin G (MIC, > 1 I1g/mL) [8].
Serotyping. The serotypes of S. pneumoniae isolates with
decreased susceptibility to penicillin G were determined at the
Centre National de Reference des Pneumocoques (Dr. Geslin,
Creteil, France) by a latex agglutination method (immune serum provided by the Statens Seruminstitut, Copenhagen).
Statistical analysis. Fisher's exact test was used for qualitative
variables, and Wilcoxon's test was used for continuous variables.
Relative risks (RRs) and 95% confidence intervals (95% CIs)
were calculated to identify which factors were most related to
PRP acquisition. Stratified analysis was then carried out with the
Mantel-Haenzel estimate. The statistical analysis was performed
with the SAS software package (SAS Institute, Cary, NC).
Results
Between January 1992 and September 1994, S. pneumoniae
was isolated from 45 individual patients. PRP were isolated
from 14 patients (group 1), and strains of S. pneumoniae that
had normal susceptibility to penicillin G were isolated from
31 patients (group 2).
In group I (table I), 8 patients (57%) had pneumonia
(3 of whom had positive blood cultures), 2 had bronchial
superinfection, 2 had sinusitis, and 2 were simply colonized.
S. pneumoniae that was highly resistant to penicillin G was
isolated from 8 patients (57%), of whom 7 had pneumonia
(blood cultures were positive for 2) and I had bronchial
superinfection. Serotype 23 was the most common, accounting for 80% of isolates in group I whose serotypes were
identified. Clinically, infection resolved regardless of resistance to penicillin G. Pneumonia was treated with amoxicillin (3-6 g/24 h; n = 4), teicoplanin (6 mg/[kg' 24 h]; n =
2), pristinamycin (3 g/24 h; n = 1), or cefotaxime (3 or 4
g/24 h; n = I). The 14 patients from whom PRP were isolated
all had a strain that was resistant to trimethoprim-sulfamethoxazole (TMP-SMZ), except for one patient who was not
receiving TMP-SMZ.
In group 2 (n = 31), 18 patients had pneumonia (12 of whom
had positive blood cultures), 9 had bronchial superinfection,
and 4 were colonized. Clinically, infection resolved in all 27
patients infected. The 31 penicillin-susceptible S. pneumoniae
isolates were susceptible to TMP-SMZ in all but 2 cases (1 of
the 2 patients was receiving TMP-SMZ).
As shown in table 2, demographic data and the distribution
of the risk factors for HIV infection were comparable between
the two groups, as were CD8 + cell counts and frequency of
qualitative p24 antigenemia. The median CD4+ cell count was
lower in the group of patients from whom a resistant strain
was isolated, although not significantly so. The distribution of
the different types of infection (pneumonia, bronchial superinfection, sinusitis) and the numbers of patients colonized were
comparable in groups I and 2, as were the numbers of patients
cm
S. pneumonia with Decreased Susceptibility to Penicillin G
1996;22 (March)
439
Table 2. Comparison of patients from whom S. pneumoniae with decreased or normal susceptibility
to penicillin G were isolated.
Susceptibility to penicillin G
Characteristics of and data
regarding patients
No. (%) of males
Age (y), median (range)
No. (%) of smokers
No. (%) addicted to drugs
Median cell count per mm 3
(range)
CD4
CD8
WBC
PMN
No. (%) with p24 antigen
No. (%) with positive
blood culture
No. (%) relapsing
During the previous 3 mo:
Admission to hospital
Treatment with
Pentamidine aerosol
Antiretroviral agents
Antibiotic(s)
,B-Lactam agents
TMP-SMZ
NOTE.
PMN
=:
Decreased
=: 14) (%)
(n
(n
10 (71)
33.5 (23-59)
12 (86)
3 (21)
22
33
22
13
25 (2-294)
391.5 (164-2,020)
4,325 (1,100-10,900)
2,100 (660-9,047)
6 (43)
48
548
5,050
3,180
11
Nonnal
=: 31) (%)
P value
(71)
(27-60)
(71)
(42)
1
.48
.45
.31
(1-807)
(30-1,822)
(1,700-15,500)
(1,224-12,400)
(35)
.09
.42
.24
.06
.74
3 (21)
3 (21)
12 (39)
1 (3)
.32
.08
3 (21)
7 (23)
.63
2
9
13
4
10
14
14
9
2
5
(14)
(64)
(93)
(29)
(71)
(45)
(45)
(29)
(6)
(16)
.09
.34
<.001
.06
<.001
polymorphonuclear neutrophil.
whose blood cultures were positive, relapse rates, and neutrophil counts at the time of infection.
Whatever the susceptibility of the strain, 20% of all patients had been admitted to a hospital in the previous 3
months. Patients from whom a resistant strain was isolated
had received more antibiotics (of all types) than had patients
from whom a susceptible strain was isolated (93% vs. 29%;
P < .001), particularly with regard to ,B-lactam agents (29%
vs. 6%; P = .06) and TMP-SMZ (71 % vs. 16%; P < .001).
The relative risks for isolation of a resistant strain associated
with administration of any antibiotic, a ,B-lactam agent, or
TMP-SMZ were 13.6 (95% CI, 1.9-95.3),2.6 (95% CI, 1.25.2), and 5 (95% CI, 1.9-13.3), respectively. The relative
risk associated with TMP-SMZ intake was not modified after
adjustment for CD4+ cell count «50/mm 3 vs. >50/mm 3 )
(adjusted RR, 4.4; 95% CI, 1.6-7) or for hospital admission
during the previous 3 months (adjusted RR, 4.9; 95% CI,
2.1-11.7).
Among the patients who had not received a ,B-lactam antibiotic (n = 39), administration of TMP-SMZ was associated
with an increased risk of infection with PRP (RR, 10.2;
95% CI, 2.6-40.6). As only six patients had previously been
treated with ,B-lactam antibiotics (of whom four had also
received TMP-SMZ), the role, if any, played by TMP-SMZ
in acquisition of a strain of PRP could not be detected.
Discussion
Because of the effects of HIV on cellular and humoral immunity, the risks of bacterial infection are increased in HIV-infected patients [1]. The respiratory tract is a primary site of
infection [2]. Many authors have highlighted the frequency of
S. pneumoniae infections and their tendency to be bacteremic
and recurrent in HIV-infected patients [2, 5, 10, 11].
Strains of S. pneumoniae with decreased susceptibility to
penicillin G have been emerging in France over recent years
[6], and HIV infection is considered a risk factor for infection
by PRP [7]. Most studies on the interaction between S. pneumoniae and HIV infection have been done in the United States,
where resistance to penicillin G is considerably less frequent
than in Europe [12]. Our study confirms the frequency ofinfection due to PRP in HIV-infected subjects. Of the 45 such
patients from whom S. pneumoniae was isolated between January 1992 and September 1994, the isolates of 14 (31 %) were
PRP. During the same period the frequency of isolation of PRP
in the general population of Saint-Antoine Hospital was 19%.
Only previous antibacterial therapy in general-and TMPSMZ or ,B-lactam therapy in particular-emerged as a risk
factor for acquisition of PRP by these HIV-infected patients.
Administration ofTMP-SMZ remained a risk factor for acquisition of PRP after adjustment for the CD4+ count and recent
(prior) admissions to a hospital.
440
Meynard et al.
Contrary to the case with ,B-lactam antibiotics [13], the
mechanism of resistance to TMP-SMZ has not been clearly
defined. It appears to be related to the production of an abnormal enzyme, dihydrofolate reductase, with abnormally high
affinity for TMP [14]. All 14 strains of PRP isolated were
resistant to TMP-SMZ, with the exception of one isolated from
a patient who was not receiving TMP-SMZ. Conversely, all
the strains susceptible to penicillin G were also susceptible to
TMP-SMZ, with the exception of two, one of which was isolated from a patient receiving TMP-SMZ. We hypothesize that
TMP-SMZ may protect against bacterial infections, particularly
those due to susceptible S. pneumoniae.
Because of their immunodeficiency, HIV-infected patients
are more likely to be infected by strains with decreased susceptibility to penicillin G, which are generally less virulent. Patients infected by HIV also have specific features ofimmunodeficiency that increase the risk of pneumonia and bacteremia
due to S. pneumoniae serotypes that are not virulent but are
often multiresistant; indeed, they have been shown to have
increased oropharyngeal colonization [15], reduced antibodydependent cytotoxic responses involving local IgA production
and alveolar macrophage activity [16], reduced IgG 2 responses
to polysaccharide antigens [17], and abnormal complement activation [18].
Another hypothesis is that administration of TMP-SMZ may
directly induce the emergence of TMP-SMZ-resistant mutants.
The genes coding for sulfonamide and TMP resistance in these
strains are physically close to the genes encoding penicillinbinding proteins 2x and 2b, which are the first to be altered in
the penicillin-resistant strains [19], and this circumstance may
explain the high frequency of resistance to sulfonamide and
TMP among these strains. Long-term use of the TMP-SMZ
combination may thus promote the selection of penicillin-resistant pneumococci.
Conclusions
This study confirms the high frequency of PRP infection and
colonization in HIV-infected patients. However, the clinical
manifestations and outcome were similar to those for patients
infected with penicillin-susceptible strains. Recent antibiotic
therapy, especially with TMP-SMZ, appears to predispose
HIV-infected patients to infection by S. pneumoniae strains
with decreased susceptibility to penicillin.
Acknowledgment
The authors are deeply indebted to Dr. P. Geslin for serotyping
strains of Streptococcus pneumoniae with decreased susceptibility
to penicillin G.
cm
1996; 22 (March)
References
1. Simberkoff MS, Leaf HL. Bacterial infections in patients with HIV infection. In: Wormser GP, ed. AIDS and other manifestations of HIV infection. New York: Raven Press, 1992:269-75.
2. Janoff EN, Breiman RF, Daley CL, Hopewell PC. Pneumococcal disease
during HIV infection: epidemiologic, clinical, and immunologic perspectives. Ann Intern Med 1992;117:314-24.
3. Polsky B, Gold JWM, Whimbey E, et al. Bacterial pneumonia in patients
with the acquired immunodeficiency syndrome. Ann Intern Med
1986; 104:38-41.
4. Witt DJ, Craven DE, McCabe WR. Bacterial infections in adult patients
with the acquired immune deficiency syndrome (AIDS) and AIDSrelated complex. Am J Med 1987; 82:900-6.
5. Pesola GR, Charles A. Pneumococcal bacteremia with pneumonia: mortality in acquired immunodeficiency syndrome. Chest 1992; 101:150-5.
6. Geslin P, Buu-Hoi A, Fremaux A, Acar JF. Antimicrobial resistance in
Streptococcus pneumoniae: an epidemiological survey in France, 19701990. Clin Infect Dis 1992; 15:95-8.
7. Bedos JP, Chastang C, Geslin P, Regnier B. Resultats de l'enquete nationale sur la resistance du pneumocoque de sensibilite diminuee aux betalactamines. In: Carbon C, Chastang C, Decazes JM, eds. Infections it
pneumocoque de sensibilite diminuee aux beta-lactamines. Paris:
Springer Verlag, 1993:121-52.
8. Acar J, Bergogne-Berezin E, Chardon H, et al. Communique 1993 du
Comite de l'Antibiogramme de la Societe Franetaise de Microbiologie.
Path Bioll993;41:741-8.
9. Jorgensen JH, Ferraro MJ, McElmeel ML, Spargo J, Swenson lM, Tenover
FC. Detection of penicillin and extented-spectrum cephalosporin resistance among Streptococcus pneumoniae clinical isolates by use of the
E test. J Clin MicrobioI1994;32:159-63.
10. Redd SC, Rutherford GW III, Sande MA, et al. The role ofhuman immunodeficiency virus infection in pneumococcal bacteremia in San Francisco
residents. J Infect Dis 1990; 162:1012-7.
II. Simberkoff MS, El Sadr W, Schiffinan G, Rahal JJ Jr. Streptococcus
pneumoniae infections and bacteremia in patients with acquired immune
deficiency syndrome, with report of a pneumococcal vaccine failure.
Am Rev Respir Dis 1984; 130:1174-6.
12. Spika JS, Facklam RR, Plikaytis BD, Oxtoby MJ, the Pneumococcal Surveillance Working Group. Antimicrobial resistance of Streptococcus
pneumoniae in the United States, 1979-1987. J Infect Dis
1991; 163:1273-8.
13. Georgopapadakou NH. Penicillin-binding proteins and bacterial resistance
to .B-lactams. Antimicrob Agents Chemother 1993;37:2045-53.
14. Klugman KP. Pneumococcal resistance to antibiotics. Clin Microbiol Rev
1990; 3: 171-96.
15. Muller F, Fmland SS, Hvatum M, Radl J, Brandtzaeg P. Both IgA subclasses are reduced in parotid saliva from patients with AIDS. Clin Exp
Immunol 1991; 83 :203 - 9.
16. Sestini P, Nencioni L, Villa L, Boraschi D, Tagliabue A. IgA-driven
antibacterial activity against Streptococcus pneumoniae by mouse lung
lymphocytes. Am Rev Respir Dis 1988; 137: 138-43.
17. Muller F, Fmland SS, Brandtzaeg P. Altered IgG-subclass distribution in
lymph node cells and serum of adults infected with human immunodeficiency virus (HIV). Clin Exp Immunol 1989;78:153-8.
18. Tausk FA, McCutchan JA, Spechko P, Schreiber RD, Gigli I. Altered
erythrocyte C3b receptor expression, immune complexes, and complement activation in homosexual men in varying risk groups for acquired
immune deficiency syndrome. J Clin Invest 1986; 78:977 -82.
19. Gasc AM, Kauc L, Baraille P, Sicard M, Goodgal S. Gene localization,
size and physical map ofthe chromosome of Streptococcus pneumoniae.
J Bacteriol 1991; 173:7361-7.