Secondary Logo

Journal Logo

Non-typhoidal salmonella bacteraemia among HIV-infected Malawian adults: high mortality and frequent recrudescence

Gordon, Melita A.a; Banda, Hastings T.a; Gondwe, Macphersona; Gordon, Stephen B.b; Boeree, Martin J.a; Walsh, Amanda L.b; Corkill, John E.c; Hart, C. Anthonyc; Gilks, Charles F.d; Molyneux, Malcolm E.b,d

Clinical Science
Free

Objective Non-typhoidal salmonella (NTS) bacteraemia is a common, recurrent illness in HIV-infected African adults. We aimed to describe the presentation and outcome of NTS bacteraemia, the pattern of recurrence, and to determine whether recurrence results from re-infection or recrudescence.

Design One hundred consecutive adult inpatients with NTS bacteraemia in Blantyre, Malawi, were treated with chloramphenicol. Survivors were prospectively followed to detect bacteraemic recurrence.

Methods Index and recurrent isolates were typed by antibiogram, pulsed-field gel electrophoresis and plasmid analysis to distinguish recrudescence from re-infection.

Results Inpatient mortality was 47%, and 1-year mortality was 77%. A total of 77 out of 78 cases were HIV positive. Anaemia was associated with inpatient death, and several features of AIDS were associated with poor outpatient survival. Among survivors, 43% (19/44) had a first recurrence of NTS bacteraemia at 23–186 days. Among these, 26% (5/19) developed multiple recurrences up to 245 days. No recurrence was seen after 245 days, despite follow-up for up to 609 days (median 214). Suppurative infections were not found at presentation, and were only seen twice at recurrence. Index and recurrent paired isolates were identical by phenotyping and genotyping, consistent with recrudescence, rather than re-infection.

Conclusion NTS bacteraemia has a high mortality (47%) and recurrence (43%) rate in HIV-infected African adults. Recurrence is caused by recrudescence rather than re-infection. As focal infections were rarely found, recrudescence may often be a consequence of intracellular tissue sequestration. There is an urgent need for improved primary treatment and secondary prophylaxis in Africa.

From the aDepartment of Medicine, University of Malawi College of Medicine, Blantyre, Malawi; bMalawi–Liverpool Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi; cDepartment of Medical Microbiology, University of Liverpool, Liverpool L69 3GA, UK; and dLiverpool School of Tropical Medicine, Liverpool L3 5QA, UK.

Reprints will not be available from the authors

Correspondence to: Dr M.A. Gordon, Wellcome Trust Research Laboratories, PO Box 30096, Blantyre, Malawi. E-mail: mgordon@mlw.medcol.mw

Received: 7 December 2001;

revised: 4 April 2002; accepted: 29 April 2002.

Sponsorship: M.E.M. was supported by a Research Leave Fellowship, and S.B.G. was supported by a Training Fellowship in Clinical Tropical Medicine, both from the Wellcome Trust, UK. This work was partly supported by a grant from the DfiD HIV Knowledge Programme. DfiD take no responsibility for the views expressed in this report.

Back to Top | Article Outline

Introduction

Non-typhoidal salmonellae (NTS) are the commonest bacterial isolates from adult blood culture series in African countries where HIV prevalence is high [1–3]. NTS comprise 37% of adult blood culture isolates in Blantyre, Malawi [4], where HIV seroprevalence is approximately 30% among urban women attending antenatal clinics [5].

In immunocompetent adults, NTS infection causes a self-limiting diarrhoeal illness. Transient bacteraemia occurs in 1–4% of cases, and overall mortality is 0.5%. By contrast, the available data suggest a devastating inpatient mortality of up to 80% among HIV-infected adults with NTS bacteraemia [4,6,7]. Recurrent NTS bacteraemia was recognized as a feature of AIDS in 1985 [8–11], but the pattern and outcome of recurrences have not been established prospectively in Africa.

Several hypotheses might explain the recurrence of NTS bacteraemia in HIV-positive African adults. First, the increased susceptibility among HIV-infected adults [12] might lead to frequent re-infections. Second, NTS may emerge repeatedly from a suppurative focus of infection. Such foci are seen in immunocompetent hosts in damaged urinary tracts, endothelium, joints or bones [13], and have been reported at more unusual sites in HIV [14–19], and these might cause recrudescence of NTS bacteraemia. In addition, schistosomiasis is responsible for the persistence and recurrence of NTS among African children [20], the tegument of adult worms providing an intravascular sanctuary site [21]. Schistosomal infections are endemic in parts of Malawi, and this might be another site from which NTS could re-emerge in an adult host. Third, salmonellae are capable of intracellular persistence in human cells of monocyte/macrophage lineage [22], and the immune deficit in HIV may permit persistence in the tissue macrophages of the reticuloendothelial system, in a manner analogous to persistence in Salmonella typhi infections.

In four Italian HIV-seropositive patients with seven recurrences of NTS bacteraemia, molecular IS200 fingerprinting was used to establish that five out of seven events were recurrence with a different organism, suggesting exogenous re-infection at 2–18 months. Two out of seven events were recurrence with the same organism, suggestive of recrudescence at one and 6 months [23]. If re-infections were also found to be the commonest cause of recurrence in Africa, then the risk of recurrent NTS infections in susceptible individuals might be reduced through behavioural strategies.

In this study we aimed first to describe the presentation of NTS bacteraemia in HIV-infected Malawian adults, and second to study prospectively the rate, timing and outcome of recurrent NTS bacteraemia in individuals surviving an index episode of NTS bacteraemia. We then used pulsed field gel electrophoresis (PFGE), and plasmid typing to establish the degree of similarity between index and recurrent organisms, and thus to distinguish new from recrudescent NTS infections.

Back to Top | Article Outline

Patients and methods

All adult general medical admissions to Queen Elizabeth Central Hospital, Blantyre, Malawi, who presented with fever or clinical evidence of septic shock, had venous blood (5 ml) taken for culture. Blood was inoculated into a single bottle of brain–heart infusion broth, and incubated at 37°C in air for 7 days, with routine sub-cultures onto sheep-blood agar in 5% carbon dioxide at 24 and 48 h and 7 days. Cloudy bottles were examined by Gram stain and subcultured onto appropriate media. One hundred consecutive patients with community-acquired NTS bacteraemia were recruited the same day a positive blood culture was reported. A standardized history and examination were recorded. Blood was taken for a full blood count, and examined for malaria parasites. HIV testing and automated CD4 cell count (Facscount; Becton Dickinson, San Jose, CA, USA) were performed for individuals who gave specific informed consent. HIV testing was performed on a single blood sample using two methods (HIV Serocard; Trinity Biotech, Ireland; and HIV-1 and 2 enzyme-linked immunosorbent assay; Ortho Clinical Diagnostics, Raritan, NJ, USA). Patients who had died before a positive blood culture result was available were included in the study, but only routine admission data and blood samples were available.

The recruiting clinician investigated any clinically apparent focus of infection, using microscopy, bacterial culture, chest X-ray, other plain radiological studies, or ultrasound examination. Empirical treatment was usually started at admission, pending blood culture results. After the identification of NTS from blood cultures, treatment with chloramphenicol was either started or continued in all patients at a minimum prescribed dose of 500 mg four times a day for 14 days. The maximum prescribed total dose of chloramphenicol was 42 g. Survivors were asked to attend a monthly follow-up clinic, and were also encouraged to attend if they developed intercurrent ill health between appointments. Transport expenses and a modest allowance were paid for each attendance.

At the first clinic attendance, 4–6 weeks after index presentation, a standardized history and examination were recorded, and compliance with treatment was assessed by examining inpatient drug charts and pharmacy dispensing records, and by direct questioning. The total dose and duration of chloramphenicol treatment, and the period since the last dose of chloramphenicol were noted. The minimum acceptable index treatment taken was considered to be 28 g total dose, and 14 days’ duration (often completed after discharge from hospital). Irrespective of clinical condition, venous blood (5 ml) for culture was taken from all patients at the first clinic attendance. Stool and urine samples from survivors were examined for ova, parasites and cysts before any anthelmintic drug was administered. Stool was prepared by formal ether concentration, and urine was centrifuged.

At all subsequent monthly visits, a systematic history and examination were recorded, and episodes of intercurrent ill health were investigated and treated, with hospital admission if necessary. At these later visits, blood culture was repeated only if patients were symptomatic or febrile. Monthly follow-up was continued for up to 20 months, or until death. Any patient who failed to attend clinic was contacted after 2 weeks to request attendance, or was visited at home by study staff.

A recurrence of NTS bacteraemia was defined as growth of NTS from blood culture after an adequate clinical response (resolution of fever and presenting symptoms) to a full course of treatment with chloramphenicol for the index infection. Positive NTS blood cultures in patients who had received an incomplete or interrupted index chloramphenicol regimen were considered to be ‘false’ recurrences. A minimum period of 5 days after the completion of treatment with chloramphenicol was required before negative blood culture was taken as evidence of clearance. All recurrences of NTS bacteraemia were managed by a clinical, radiological and bacteriological search for localized foci of infection, and by re-treatment with chloramphenicol at a dose of 250 mg four times a day for a period of 28 days, after which the patient was re-assessed and blood culture was repeated.

Pairs or series of index and recurrent NTS blood culture isolates were tested for antibiotic susceptibility by disc diffusion. PFGE of macrorestricted chromosomal DNA was performed as follows: bacteria were harvested from solid media, embedded in lysozyme-containing agarose, and lysed using a bile-detergent buffer. Released nucleases were neutralized with proteinase K. Chromosomal DNA was digested with the restriction endonuclease Spe 1 (30 units per plug), and digestion fragments were separated by PFGE (CHEF 3D; BioRad Ltd., Hercules, CA, USA). PFGE conditions were as follows: buffer temperature 14°C, 6 volts/cm, pulsing angle 120°, initial switch time 5 s, with a linear increase to 33 s over 20 h. The relatedness of strains on PFGE was assessed according to the Tenover criteria [24].

Strains that could not be differentiated using these methods were further investigated by plasmid typing, performed by standard alkaline lysis and agarose gel electrophoresis.

Back to Top | Article Outline

Statistical analysis

Data were prospectively collected on standard proformas, entered in an MS Access database, and analysed using stata 6.0 (Stata Corporation, Texas, USA).

The association of inpatient death (binary outcome) with clinical and laboratory presenting features was first tested in univariate analyses. χ2 tests were used for binary variables, and t-tests or rank-sum tests for continuous variables. Any association reaching a significance of P < 0.1 in these univariate analyses was then tested for interaction, and corrected multivariate Mantel–Haenszel odds ratios (OR) were calculated.

The association of outpatient recurrence and outpatient survival (time series data) with clinical and laboratory presenting features were tested using Cox's proportional hazard regression analysis. Kaplan–Meier survival estimates were plotted for time to first recurrence and for overall survival.

Back to Top | Article Outline

Ethical approval

This study was given ethical approval by the University of Malawi College of Medicine Research Committee.

Back to Top | Article Outline

Results

Recruitment

Recruitment is summarized in Fig. 1. One hundred patients (54 men) were recruited at median day 2 (range 1–5). The median age was 32 years (range 18–60), 59 lived in urban townships and 28 in rural areas. Two patients had previously been inpatients at Queen Elizabeth Central Hospital, with confirmed NTS bacteraemia.

Fig. 1.

Fig. 1.

Back to Top | Article Outline

Clinical features at index presentation

The index clinical presentation is summarized in Table 1. There was no clinical evidence of infective endocarditis in any case, and no patient had evidence of gastrointestinal bleeding or perforation during admission.

Table 1

Table 1

Back to Top | Article Outline

Microbiological findings

NTS blood isolates from the 100 cases consisted of Salmonella typhimurium (75), Salmonella enteritidis (19), a dual growth of S. typhimurium and S. enteritidis (one), and other Salmonella spp. (five). Admission blood culture grew two pathogens from the same sample in two other patients (S. typhimurium with Klebsiella spp., S. typhimurium with Streptococcus pneumoniae). Bacterial susceptibility testing overall showed 5% resistance to chloramphenicol, 73% to co-trimoxazole, 79% to ampicillin, 43% to gentamycin and 40% to tetracycline.

The investigation of clinically apparent foci of infection in 47 patients included 23 Ziehl–Neilsen sputum examinations for acid and alcohol-fast bacilli, the culture of five transthoracic aspirates, and microscopy and culture of eight pleural aspirates, 10 cerebrospinal fluid examinations, and one each of joint, skin pustule and ascitic aspirates. These investigations identified pathogens in 11 patients. These consisted of four patients with respiratory bacterial co-infections, with S. pneumoniae (one), klebsiella (two) and citrobacter (one), two patients with acid and alcohol-fast bacilli-positive pulmonary tuberculosis, three with cryptococcal meningitis, and two with Staphylococcus aureus focal infections (one skin, one joint). Pleural effusions in two patients were attributed to tuberculosis (lymphocytic effusion with good response to treatment), and pleural effusions in two patients and ascites in one patient were attributed to widespread Kaposi's sarcoma. There was therefore no evidence of localized or suppurative Salmonella infection in any case at presentation.

Back to Top | Article Outline

Laboratory findings

Full blood counts (available for 70 patients) showed median haemoglobin 6.8 g/dl (2.5–11.7), median white cell count 3.8 × 109/l (1–17), and median platelet count 104 × 109/l (14–406). Malaria parasites were seen in 12 out of 70 samples (17%). Missing laboratory data were predominantly from patients who died within 48 h. Stool microscopy showed helminths in seven out of 40 surviving patients, including Schistosoma mansoni in two.

Back to Top | Article Outline

HIV and CD4 cell testing

Seventy-seven out of 78 patients (99%) tested were seropositive for HIV by both methods. CD4 cell counts were performed on 59 patients at presentation or at first follow-up, or both. The median CD4 cell count among HIV-positive patients at presentation was 99 cells/μl (range 6–313, n = 50), and at follow-up it was 108 cells/μl (range 6–445, n = 34). The highest recorded count (median 101 cells/μl) was taken as the best estimate of the baseline CD4 cell count.

Back to Top | Article Outline

Inpatient treatment and outcome

In total, 93 out of 100 patients received chloramphenicol, which was commenced at median day 0 (0–15). Empirical antibiotic treatment was given to 85 patients on day 0, and included chloramphenicol for 49 patients. A further 44 patients were commenced on chloramphenicol when blood culture results became available. Chloramphenicol treatment was delayed beyond day 5 in two patients who absconded, and were recalled. Forty-seven out of 100 patients died within one month as inpatients, at a median of 4 days (range 0–30). The median length of admission among survivors was 5 days (range 0–26).

Back to Top | Article Outline

Follow-up visits

A total of 48 patients attended the follow-up clinic, where compliance with medication was assessed. The median total dose of chloramphenicol taken was 32 g (range 10–42), and the median duration of treatment was 14 days (range 4–18). Forty-four out of 48 survivors completed the minimum index treatment, and four took incomplete or interrupted courses.

A total of 410 follow-up visits were made (312 scheduled and 98 unscheduled), and there were 156 intercurrent clinical episodes, including 29 re-admissions to hospital. The commonest intercurrent events were acute diarrhoea (40), acute lower respiratory infection (34), acute non-specific febrile illness (34), bacterial or fungal skin infections (nine), painful peripheral neuropathy (eight), and pulmonary tuberculosis (six).

Back to Top | Article Outline

Follow-up blood cultures and recurrences

A total of 48 patients had 145 blood cultures taken during follow-up, and 42 out of 145 cultures (29%) grew a bacterial pathogen. Twenty-one patients had blood cultures positive for NTS on 31 occasions. Eight patients had positive blood cultures for other pathogens on 11 occasions (seven S. pneumoniae, four others). Three of these eight individuals were among the 21 who also had a recurrence of NTS bacteraemia. Of these 21 individuals, two had a ‘false’ recurrence after incomplete index treatment. Nineteen patients had a ‘true’ recurrence of NTS bacteraemia on 29 occasions, and were re-treated with chloramphenicol.

Back to Top | Article Outline

Detection, timing and outcome of non-typhoidal salmonella recurrence

The follow-up, detection and outcome of recurrences are summarized in Fig. 1. Eleven out of 44 subjects demonstrated true recurrence of NTS bacteraemia at the first clinic attendance. The period since the last dose of chloramphenicol was a median of 18 days (range 8–35). Although blood culture was performed regardless of clinical state, these 11 were all symptomatic or febrile. One patient had developed an S. typhimurium urinary infection, which was successfully treated, but the patient went on to have a second recurrence of bacteraemia without focus at 100 days. The remaining 10 had no evidence of suppurative infection at first recurrence. Another of these 10 also had multiple recurrences of NTS bacteraemia, and developed a new purulent thoracic empyema at the second recurrent event, from which S. typhimurium was isolated.

As the follow-up progressed, eight individuals developed later first recurrence of NTS bacteraemia, after a negative blood culture at the first visit. Three developed multiple events of recurrent NTS bacteraemia, but there was no evidence of suppurative infection among these eight patients at any recurrent event.

The overall timing of recurrences of NTS bacteraemia is shown in Fig. 2b. No first NTS recurrence was detected after 186 days, and no multiple events were detected after 245 days, despite the follow-up of cases to a maximum of 609 days (median 214 days).

Fig. 2.

Fig. 2.

Back to Top | Article Outline

Clinical features of first recurrence

The clinical features of the 19 first events of recurrent NTS bacteraemia are summarized in Table 1. Overall, there were fewer symptoms and signs at recurrence compared with index presentation, and fever without identifiable clinical focus became the commonest presentation. Five out of 19 patients (26%) died within 1 month of first recurrence, a lower immediate mortality rate than at the index presentation.

Back to Top | Article Outline

Features associated with inpatient death and poor survival

The Kaplan–Meier estimate of overall survival from the day of admission is shown in Fig. 2a. Mortality at one year was 77%. The presenting features associated with inpatient death and poor outpatient survival are presented in Table 2. Inpatient death was associated with reduced admission haemoglobin concentration and reduced Glasgow Coma Score (GCS) on univariate analysis. The haemoglobin concentration when corrected for GCS remained highly significant as a predictor of inpatient death [OR 1.6 for each fall of 1 g of haemoglobin concentration, 95% confidence interval (CI) 1.3–2.0;P = 0.0001].

Table 2

Table 2

Poor outpatient survival was associated with several features of AIDS on univariate Cox's regression analysis. The recurrence of NTS bacteraemia was not associated with any clinical or laboratory feature of presentation, and recurrence did not affect survival as an outpatient in this study. The NTS serovar and antibiotic resistance pattern were not associated with poor inpatient or outpatient outcome, or with recurrence.

Only one patient was HIV negative, an elderly man who was followed to 600 days, with no recurrence or other intercurrent illness.

Back to Top | Article Outline

Phenotypic and genotypic analysis of recurrent isolates

Fifteen pairs or series of index and recurrent isolates were available for analysis (summarized in Table 3). Serology, antibiogram and PFGE showed six out of 15 subjects to have unique index organisms. Plasmid analysis allowed the remaining nine subjects to be further split into three distinct groups. There were therefore nine different index organisms among 15 patients, representing considerable inter-individual heterogeneity.

Table 3

Table 3

All 15 pairs of isolates showed intra-individual concordance at the first recurrence. There were three series of isolates available for analysis, two of which showed concordance. Patient E was the only subject in whom there was evidence of re-infection with a non-concordant strain, which occurred at the third recurrence.

Back to Top | Article Outline

Discussion

This is the largest description of the clinical course of this common HIV-related condition in African adults; 47% died in hospital within one month, 77% had died after one year, and 43% of outpatient survivors developed a recurrence of NTS bacteraemia on one or more occasions. Using molecular techniques, we have demonstrated that these recurrences were caused by isolates that were the same as the original infecting organism. NTS bacteraemia is a severe, recrudescent disease.

A reduction in haemoglobin concentration was strongly associated with inpatient death; all the patients in the study were anaemic, 50% being severely anaemic with haemoglobin concentrations below 7 g/dl. Normochromic normocytic anaemia is a multifactorial feature of advanced HIV disease, and merits further prospective study as a contributor to early death. Several features of AIDS were associated with poor outpatient survival.

A recurrence of NTS bacteraemia was reported as a feature of AIDS in 1985; we confirm its importance in African HIV disease, and describe the timing and pattern for the first time. NTS comprised 27% of positive blood cultures among medical admissions, but 72% of positive blood cultures during follow-up. At least one recurrence of NTS bacteraemia was found in 43% of adequately treated survivors, and 26% went on to have multiple recurrences. Lower rates of first recurrence (five out of 23) were reported from a community cohort in Kenya who had higher CD4 cell counts at presentation [25], but the CD4 cell count did not predict recurrence in our study.

All patients with recurrence detected at the first clinic visit were symptomatic or febrile, indicating that we did not merely document incidental bacteraemia. Only 5 ml of blood were taken for culture, so our data may under-represent the true rate of recurrence. Recurrence presented with fewer symptoms and signs, had a lower case fatality than the index illness, and did not appear to affect long-term outpatient survival. Recurrence may be intrinsically a milder clinical event than the first episode of bacteraemia, but this study involved active follow-up and re-treatment; the course of late-presenting recurrences might be very different. Only two out of 100 of our self-presenting sample had a previous admission for confirmed NTS bacteraemia. This might reflect the apparent mildness of recurrence, but could equally well reflect inadequate investigation of previous illness.

PFGE and plasmid typing are well-established tools to distinguish between different strains of salmonellae [26,27], and showed in this study that whereas isolates differed considerably between patients, all patients had first recurrences with an organism identical to the index isolate. Only one showed evidence of re-infection at a late recurrent event. Taken together, these data make recrudescence the most likely explanation for the high recurrence rate in this cohort, particularly in view of the contrasting findings in the Italian study, in which five out of seven recurrent isolates were different from the original organism [23]. Several factors could have contributed to the high rate of recrudescence. Case reports have described suppurative salmonella infections in the chest and elsewhere in HIV disease, and many of our patients had respiratory symptoms and signs. Focal NTS infections were carefully sought, but were not found in any case at index presentation, and were found in only two out of 19 patients at the time of recurrence. In contrast, focal co-infections with other pathogens were identified in 11 out of 100 patients at index presentation. Focal or suppurative NTS infections therefore appeared to contribute to recrudescence in a minority of cases only. Similarly, schistosomal ova were found in stool in only two out of 40 patients. Although HIV infection may reduce the egg count, and more sensitive methods might have identified more cases, the very low numbers detected make it unlikely that schistosomal co-infection played a major role in causing NTS recurrence. It therefore seems likely that in many cases NTS have an alternative sanctuary site within the HIV-infected human host. By analogy with typhoid fever, this may be intracellular within macrophages in reticuloendothelial tissues.

Intracellular survival is a key element in salmonella pathogenesis, and there is an association of serious NTS infections with rare human genetic defects of the type 1 cytokine pathway [28], particularly of the IL-12 or IL-12R genes [29,30]. The deficits of type 1 cytokines described in HIV infection [31,32] may permit intracellular persistence, and underlie the observed susceptibility to severe invasive and recurrent NTS infections.

Our findings carry implications for the management of these cases. Chloramphenicol failed to effect a radical cure of NTS bacteraemia, and the strategy of re-treatment with chloramphenicol for recurrence was also only partly successful. Index treatment with fluoroquinolones or third-generation cephalosporins might reduce inpatient mortality or recurrence. The present rising tide of multidrug-resistant NTS in Malawi and elsewhere in Africa [33] will mean that these more expensive agents may soon become the only effective treatment. In affluent countries, a long-term fluoroquinolone suppressive regimen is commonly used to prevent recurrence [34]. Chloramphenicol is too toxic to be useful for long-term treatment, and fluoroquinolones are too costly in Africa. Co-trimoxazole may be useful for secondary suppression in areas where susceptibility is high, but where susceptibility is low, including Malawi, re-treatment may remain the only financially viable option. The low rate of re-infection suggests that secondary hygiene advice to avoid re-infection after an index presentation is unlikely to be helpful.

Although improved treatment regimens might improve outcome, the scarcity of microbiological facilities, the increase in multidrug-resistant strains of NTS worldwide, and the high cost of effective antimicrobial agents will continue to make this a difficult infection to diagnose and treat effectively in the immediate future in Africa.

Back to Top | Article Outline

Acknowledgements

The authors extend their grateful thanks to the staff and patients of the Department of Medicine (University of Malawi College of Medicine), and to the staff of the Main Laboratory at Queen Elizabeth Central Hospital, Blantyre, for their generous participation and co-operation.

Back to Top | Article Outline

References

1. Gilks CF, Brindle RJ, Otieno LS. et al. Life-threatening bacteraemia in HIV-1 seropositive adults admitted to hospital in Nairobi, Kenya. Lancet 1990, 336: 545–549.
2. Archibald LK, den Dulk MO, Pallangyo KJ, Reller LB. Fatal Mycobacterium tuberculosis bloodstream infections in febrile hospitalized adults in Dar es Salaam, Tanzania. Clin Infect Dis 1998, 26: 290–296.
3. Vugia DJ, Kiehlbauch JA, Yeboue K. et al. Pathogens and predictors of fatal septicemia associated with human immunodeficiency virus infection in Ivory Coast, West Africa. J Infect Dis 1993, 168: 564–570.
4. Gordon MA, Walsh AL, Chaponda M. et al. Bacteraemia and mortality among adult medical admissions in Malawi – predominance of non-typhi Salmonellae and Streptococcus pneumoniae. J Infect 2001, 42: 44–49.
5. Taha TE, Dallabetta GA, Hoover DR. et al. Trends of HIV-1 and sexually transmitted diseases among pregnant and postpartum women in urban Malawi. AIDS 1998, 12: 197–203.
6. Arthur G, Nduba VN, Kariuki SM, Kimari J, Bhatt SM, Gilks CF. Trends in bloodstream infections among human immunodeficiency virus-infected adults admitted to a hospital in Nairobi, Kenya, during the last decade. Clin Infect Dis 2001, 33: 248–256.
7. Thamlikitkul V, Dhiraputra C, Paisarnsinsup T, Chareandee C. Non-typhoidal Salmonella bacteraemia: clinical features and risk factors. Trop Med Intern Health 1996, 1: 443–448.
8. Smith PD, Macher AM, Bookman MA. et al. Salmonella typhimurium enteritis and bacteremia in the acquired immunodeficiency syndrome. Ann Intern Med 1985, 102: 207–209.
9. Nadelman RB, Mathur-Wagh U, Yancovitz SR, Mildvan D. Salmonella bacteremia associated with the acquired immunodeficiency syndrome (AIDS). Arch Intern Med 1985, 145: 1968–1971.
10. Casado JL, Valdezate S, Calderon C. et al. Zidovudine therapy protects against Salmonella bacteremia recurrence in human immunodeficiency virus-infected patients. J Infect Dis 1999, 179: 1553–1556.
11. Jacobs JL, Gold JW, Murray HW, Roberts RB, Armstrong D. Salmonella infections in patients with the acquired immunodeficiency syndrome. Ann Intern Med 1985, 102: 186–188.
12. Levine WC, Buehler JW, Bean NH, Tauxe RV. Epidemiology of nontyphoidal Salmonella bacteremia during the human immunodeficiency virus epidemic. J Infect Dis 1991, 164: 81–87.
13. Cohen JI, Bartlett JA, Corey GR. Extra-intestinal manifestations of Salmonella infections. Medicine (Baltimore) 1987, 66: 349–388.
14. Aliaga L, Mediavilla JD, Lopez DO, Lopez-Gomez M, de Cueto M, Miranda C. Nontyphoidal salmonella intracranial infections in HIV-infected patients. Clin Infect Dis 1997, 25: 1118–1120.
15. Casado JL, Navas E, Frutos B. et al. Salmonella lung involvement in patients with HIV infection. Chest 1997, 112: 1197–1201.
16. Fernandez GM, Ramos JM, Nunez A, de Gorgolas M. Focal infections due to non-typhi Salmonella in patients with AIDS: report of 10 cases and review. Clin Infect Dis 1997, 25: 690–697.
17. Fernandez GM, Torres PR, Gomez RJ, Nunez GA, Jusdado JJ, Ramos RJ. Infectious endocarditis due to non-typhi Salmonella in patients infected with human immunodeficiency virus: report of two cases and review. Clin Infect Dis 1996, 22: 853–855.
18. Medina F, Fuentes M, Jara LJ, Barile L, Miranda JM, Fraga A. Salmonella pyomyositis in patients with the human immunodeficiency virus. Br J Rheumatol 1995, 34: 568–571.
19. Collazos J, Mayo J, Martinez E, Blanco MS. Muscle infections caused by Salmonella species: case report and review. Clin Infect Dis 1999, 29: 673–677.
20. Gendrel D, Kombila M, Beaudoin-Leblevec G, Richard-Lenoble D. Nontyphoidal salmonellal septicemia in Gabonese children infected with Schistosoma intercalatum. Clin Infect Dis 1994, 18: 103–105.
21. Melhem RF, LoVerde PT. Mechanism of interaction of Salmonella and Schistosoma species. Infect Immun 1984, 44: 274–281.
22. Buchmeier NA, Heffron F. Intracellular survival of wild-type Salmonella typhimurium and macrophage-sensitive mutants in diverse populations of macrophages. Infect Immun 1989, 57: 1–7.
23. Rubino S, Spanu L, Mannazzu M. et al. Molecular typing of non-typhoid Salmonella strains isolated from HIV-infected patients with recurrent salmonellosis. AIDS 1999, 13: 137–139.
24. Tenover FC, Arbeit RD, Goering RV, Mickelson PA, Murray BE, Persing DH. Interpreting chromosomal DNA restriction patterns produced by pulsed field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995, 33: 2233–2239.
25. Gilks CF, Ojoo SA, Ojoo JC. et al. Invasive pneumococcal disease in a cohort of predominantly HIV-1 infected female sex-workers in Nairobi, Kenya. Lancet 1996, 347: 718–723.
26. Kariuki S, Gilks C, Kimari J, Muyodi J, Waiyaki P, Hart CA. Analysis of Salmonella enterica serotype Typhimurium by phage typing, antimicrobial susceptibility and pulsed-field gel electrophoresis. J Med Microbiol 1999, 48: 1037–1042.
27. Kariuki S, Cheesbrough J, Mavridis AK, Hart CA. Typing of Salmonella enterica serotype paratyphi C isolates from various countries by plasmid profiles and pulsed-field gel electrophoresis. J Clin Microbiol 1999, 37: 2058–2060.
28. Ottenhoff TH, Kumararatne D, Casanova JL. Novel human immunodeficiencies reveal the essential role of type-I cytokines in immunity to intracellular bacteria. Immunol Today 1998, 19: 491–494.
29. de Jong R, Altare F, Haagen IA. et al. Severe mycobacterial and Salmonella infections in interleukin-12 receptor-deficient patients. Science 1998, 280: 1435–1438.
30. Altare F, Lammas D, Revy P. et al. Inherited interleukin 12 deficiency in a child with bacille Calmette–Guerin and Salmonella enteritidis disseminated infection. J Clin Invest 1998, 102: 2035–2040.
31. Chehimi J, Starr SE, Frank I. et al. Impaired interleukin 12 production in human immunodeficiency virus-infected patients. J Exp Med 1994, 179: 1361–1366.
32. Subauste CS, Wessendarp M, Smulian AG, Frame PT. Role of CD40 ligand signaling in defective type 1 cytokine response in human immunodeficiency virus infection. J Infect Dis 2001, 183: 1722–1731.
33. Hart CA, Beeching NJ, Duerden BI. et al. Infections in AIDS. J Med Microbiol 2000, 49: 947–967.
34. Hohmann EL. Nontyphoidal salmonellosis. Clin Infect Dis 2001, 32: 263–269.
Keywords:

Africa; bacterial infection; HIV/AIDS; non-typhoidal salmonella; plasmid; pulsed-field gel electrophoresis; recurrence

© 2002 Lippincott Williams & Wilkins, Inc.