Share this article on:

Brief Report: Randomized Controlled Trial of Zinc Supplementation for Persistent Diarrhea in Adults With HIV-1 Infection

Cárcamo, César MD, PhD*†; Hooton, Thomas MD‡§; Weiss, Noel S MD, DrPH*; Gilman, Robert MD‖¶; Wener, Mark H MD; Chavez, Victor MD#; Meneses, Rosario MD, PhD**; Echevarria, Juan MD††; Vidal, Margot MD**; Holmes, King K MD, PhD*‡§

JAIDS Journal of Acquired Immune Deficiency Syndromes: October 1st, 2006 - Volume 43 - Issue 2 - p 197-201
doi: 10.1097/01.qai.0000242446.44285.b5
Brief Report: Clinical Science

Background: In children, zinc supplementation reduces the incidence and severity of diarrhea.

Methods: HIV-infected adults with ≥7 days of diarrhea recruited at 3 tertiary hospitals in Lima, Peru, received a zinc sulfate capsule containing 50 mg of elemental zinc twice daily or an identical placebo for 14 days. Outcomes included persistence of diarrhea at day 14 and time until cessation of diarrhea.

Results: The 81 subjects randomized to zinc and 78 randomized to placebo were comparable at baseline, except for higher prevalences of certain enteric pathogens in the zinc group; complete follow-up rates were 62% and 69%, respectively. Zinc concentrations were consistent with zinc deficiency at follow-up in 94% of placebo recipients and 66% of zinc recipients (P = 0.01). Persistence of diarrhea at day 14 according to follow-up interview (60% for zinc-treated patients and 57.4% for placebo-treated patients) or to patient diary (42.2% vs. 31.9%) did not differ significantly. Adjusting for enteric pathogens and CD4 count, the hazard ratio (HR) for zinc supplementation and cessation of diarrhea (according to the diaries) was 0.91 (95% confidence interval [CI]: 0.50 to 1.64).

Conclusion: Supplemental zinc had no significant effect on the duration or remission of diarrhea in HIV-infected adults.

Received for publication January 13, 2005; accepted June 27, 2006.

From the *Department of Epidemiology, University of Washington, Seattle, WA; †School of Public Health and Administration, Universidad Peruana Cayetano Heredia, Lima, Peru; ‡Department of Medicine, University of Washington, Seattle, WA; §Center for AIDS and STD, University of Washington, Seattle, WA; ‖Department of Microbiology, Universidad Peruana Cayetano Heredia, Lima, Peru; ¶School of Hygiene and Public Health, Johns Hopkins University, Baltimore, MD; #Dos de Mayo Hospital, Lima, Peru; **Loayza Hospital, Lima, Peru; and ††Alexander Von Humboldt Tropical Medicine Institute, Universidad Peruana Cayetano Heredia, Lima, Peru.

Supported by the Fogarty International Center International AIDS Research Training Program (IARTP) grant D43 TW00007-06, University of Washington Center for AIDS Research, University of Washington School of Medicine Anderson Foundation, and Centers for Disease Control and Prevention.

Reprints: King K. Holmes, MD, PhD, Harborview Medical Center, Box 359931, 325 9th Avenue, Seattle, WA 98104 (e-mail:

Antimicrobial treatment of persistent diarrhea in persons with AIDS remains problematic, especially in developing countries. Diarrhea with Cryptosporidium spp. often proves refractory, Salmonella infections often recur, and diarrhea not associated with demonstrable bacterial or parasitic pathogens often persists despite antimicrobial or symptomatic therapy.

Zinc deficiency, which is found in 30% to 51% of HIV-infected patients,1-4 has been associated with impaired immunity 5 and with microsporidia infections and pathogen-negative diarrhea.6 Dietary zinc supplementation in zinc-deficient patients improves markers of cellular immune status.7 Several trials in children have demonstrated the effectiveness of zinc supplementation in the prevention and treatment of acute and persistent diarrhea.8,9 One trial demonstrated safety and efficacy in preventing watery diarrhea in HIV-infected children.10

This present trial was undertaken to assess the effect of dietary zinc supplementation on the persistence and severity of diarrhea in HIV-infected adults. Because nutritional intake required to attain normal levels of zinc seems higher in HIV-infected patients than in immunocompetent populations,11 we elected to administer 100 mg of zinc daily, more than 5 times the recommended daily allowance (RDA), and assessed the impact of this on plasma zinc concentrations. Inflammation, infection, or tissue necrosis has also been associated with low zinc levels, which are attributable to zinc redistribution within the tissues of the body, increased zinc excretion, or both.12,13 Because elevations of the acute-phase reactant, C-reactive protein (CRP), in patients with falciparum malaria14 and with tuberculosis15 have been associated with low plasma zinc levels, we assessed the relation between plasma zinc and CRP in participants.

Back to Top | Article Outline


Participants in this trial of supplemental zinc versus placebo for persistent diarrhea were participants in a case-control study of the microbial etiology and manifestations of persistent diarrhea in HIV-seropositive persons in Lima, Peru, between June 1998 and January 2000.16 Cases represented HIV-infected adult men and women with diarrhea lasting for ≥7 days and without prior treatment. Medifarma Pharmaceutical Laboratories in Lima, Peru, prepared the zinc and placebo in gelatin capsules. Standardized interviews assessed symptoms of diarrhea and HIV infection. Participants provided blood in anticoagulated trace element-free tubes and 3 fresh stool samples.

Those assigned to zinc received 30 capsules, with each containing zinc sulfate heptahydrate equivalent to 50 mg of elemental zinc. Those assigned to placebo received 30 capsules identical to the zinc capsules containing only excipients. Participants were prescribed 2 capsules once daily at 11 AM, at least 1 hour before lunch, for 2 weeks. Computer-generated randomized complete blocks, each consisting of 2 zinc and 2 placebo assignments, determined the sequence of treatment assignment. The assignment roll was kept in a locked file not accessible to personnel distributing study capsules. Participants and interviewers were blinded to the assignment arm. A code permitted the identification of the treatment group only after study completion. Participants completed daily records of diarrhea symptoms, including frequency of bowel movements and form of stools for 2 weeks after the baseline visit, and recorded the form of the stools by choosing between 5 published illustrations.17

Participants' physicians followed existing guidelines in prescribing antimicrobial treatment for those with enteric bacterial pathogens, usually sulfamethoxazole-trimethoprim (SMX-TMP) but occasionally ciprofloxacin; several patients did not fill these prescriptions. At a second visit scheduled 2 weeks after beginning treatment, interviewers assessed changes in therapy, symptoms related to diarrhea, or possible adverse effects of treatment. Remaining capsules were counted, diaries were recovered, and a second blood specimen was obtained in a trace element-free anticoagulated tube.

As previously reported, stools were tested for intestinal parasites, bacterial pathogens, Clostridium difficile toxin A, and rotavirus.16 Initial blood specimens were tested for white blood cell and CD4 lymphocyte counts, for plasma zinc and copper levels (as the average of duplicates, measured by atomic absorption on a Unicam Solaar 989 Atomic Absorption Spectrometer [Thermo Electron Corp., Waltham, MA]), and for CRP levels by nephelometry. For zinc and copper, concentrations of 85 μg/dL or greater were defined as normal, from 75 to 84 μg/dL as marginal deficiency, and less than 75 μg/dL as overt deficiency.18 Stool tests and CD4 cell count results were provided within 2 to 10 days to participants and their physicians. Institution Review Boards (IRBs) of the University of Washington, Cayetano Heredia University, Dos de Mayo Hospital, and PRISMA (a local nongovernmental organization [NGO]) approved the study protocol. Participants gave written informed consent.

Effects of zinc supplementation were assessed by “intent-to-treat” analyses. A sample size of 83 per arm was originally estimated to detect a decline from 50% to 25% in the proportion of participants reporting persisting diarrhea at follow-up, defined as at least 3 semiformed or liquid stools recorded on day 14 or 15 after starting zinc or placebo, with α = 5% and power = 80% and assuming 20% attrition. The χ2 test, Fisher exact test, and Student t test were used in univariate analyses; unconditional logistic regression was used in multivariate models to adjust for covariates, and Cox regression was used to analyze time to cessation of diarrhea. The diary was also used to determine the duration of diarrhea after enrollment by identifying the last day of follow-up when at least 3 liquid or semiliquid stools (forms 3, 4, and 5 in the diary) were recorded.

Back to Top | Article Outline


Of 159 participants randomized, 81 received zinc and 78 received placebo. Deficiency was more common for zinc (85%) than for copper (12%), which are frequencies similar to those found in HIV-seropositive controls without diarrhea from the parent case-control study (86% and 6%, respectively).16 Zinc and placebo recipients had similar demographic characteristics, duration of diarrhea, CD4 cell count, proportion with CD4 cell counts <200 cells/mm3, and plasma levels of zinc and copper. Several enteric pathogens were detected more frequently in the zinc arm (Table 1), with >1 enteric pathogen found in 19 zinc and 8 placebo recipients.



Overall, 65.4% of zinc recipients and 70.5% of placebo recipients returned for a second visit. The average number of capsules returned at 2 weeks was 12.8 for the zinc arm versus 7.5 for the placebo arm (P = 0.02), suggesting that the placebo was more acceptable or better tolerated. Participants attributing gastrointestinal symptoms to the medication included 11 of 53 given zinc versus 6 of 55 given placebo (P = 0.19). Marginal or overt zinc deficiency at follow-up was observed in 65.6% of participants given zinc versus 93.7% given placebo (P = 0.01). Zinc supplementation did not induce copper deficiency.

Of 104 participants completing follow-up questionnaires, 60% given zinc and 57% given placebo reported persisting diarrhea (Table 2). Persisting diarrhea was reported less frequently in diaries than at interview for both arms; however, again, these frequencies did not differ significantly between the 2 arms. Other gastrointestinal symptoms and fever during the second week also were reported with similar frequency in the 2 arms. Cessation of diarrhea as recorded in the diary was nonsignificantly delayed in the zinc group compared with the placebo group (hazard ratio [HR] for cessation = 0.91, 95% confidence interval [CI]: 0.50 to 1.64). Adjusting for CD4 count or for detection of a pathogen in the stool did not change the HR significantly (Table 3; Fig. 1). Similarly, stratifying for duration of diarrhea, CD4 cell count, and plasma zinc levels at enrollment disclosed no significant differences between the zinc and placebo arms in subsequent persistence of diarrhea. No individual pathogen significantly predicted diarrhea persistence, although independent of treatment arm, isolation of an enteric bacterial pathogen was associated with persistence of diarrhea as recorded in diaries (adjusted odds ratio [OR] = 3.51, 95% CI: 1.2 to 10.3). Among those with a bacterial pathogen, the HR for diarrhea cessation was even lower for the zinc arm than for the placebo arm (HR = 0.49, 95% CI: 0.1 to 2.1; ie, diarrhea lasted nonsignificantly longer in the zinc arm). Among patients who returned diaries, antibiotic susceptibility tests indicated that only 1 of 15 patients with pathogenic enteric bacterial infection received adequate doses of antibiotics active against the pathogen found and that only 1 of 11 patients with Giardialamblia infection received metronidazole. (Physicians had access to laboratory results, but some patients did not return for laboratory test results and others apparently did not obtain the metronidazole prescribed. At the time of the study, the local Ministry of Health only provided SMX-TMP free to patients with HIV infection.)







Among all cases with diarrhea, baseline zinc levels were nonsignificantly lower with higher CRP levels (P = 0.13). On pooling data from all cases with diarrhea and the HIV-infected controls without diarrhea,16 baseline zinc levels were negatively associated in a quadratic model with baseline CRP levels ([Zn] (mg/dL of plasma) = 71.7 − 2.25 × CRP (mg/dL of plasma) + 0.11 × (CRP)2; P = 0.02).

Back to Top | Article Outline


A 2-week course of 100 mg of elemental zinc daily had no significant effect on the persistence of diarrhea in HIV-infected adults. This study had limitations. Although physicians followed existing guidelines for treating diarrhea, many bacterial pathogens were resistant to SMX-TMP.16 Most G. lamblia infections also remained untreated. Even among the subset of patients with no easily treatable pathogen identified (most in this study), zinc supplementation had no significant effect. Loss to follow-up was relatively high but similar between the 2 treatment groups. Longer treatment or follow-up beyond 2 weeks might have revealed treatment benefit. Among children, however, benefits of zinc supplementation on diarrhea have been evident after the fourth day of supplementation.19

As reviewed elsewhere,3 several randomized trials have demonstrated that zinc supplementation in children reduces the incidence, duration, and severity of diarrhea and stool output; improves intestinal permeability; and prevents weight loss or stunted growth secondary to diarrhea. Among adults with AIDS-related diarrhea-wasting syndrome, however, a randomized controlled trial in Zambia demonstrated no effect of multiple micronutrient supplementation that included zinc (200 mg/d for 14 days) to improve the response to albendazole.20 Plasma levels of zinc were not measured. Differences in reported responses to zinc between children with diarrhea and adults with HIV infection and diarrhea could reflect age-dependent differences in zinc homeostasis, host immune response, inclusion criteria, or supplementation regimen, or they could reflect the possibility that low plasma concentrations of zinc in persons with HIV infection do not reflect true zinc deficiency. Future research on HIV-associated diarrhea in adults in developing countries should assess more potentially effective antimicrobial interventions to inform guidelines for empiric antimicrobial treatment algorithms.

Back to Top | Article Outline


The authors thank Medifarma Pharmaceutical Laboratories for preparing zinc and placebo capsules and the US Navy Medical Research Center Detachment, Lima, Peru, for assistance and support for this study.

Back to Top | Article Outline


1. Bogden J, Baker H, Frank O, et al. Micronutrient status and human immunodeficiency virus (HIV) infection. Ann NY Acad Sci. 1990;587:189-195.
2. Beach RS, Mantero-Atienza E, Shor-Posner G, et al. Specific nutrient abnormalities in asymptomatic HIV-1 infection. AIDS. 1992;6:701-708.
3. Koch J, Neal EA, Schlott MJ, et al. Zinc levels and infections in hospitalized patients with AIDS. Nutrition. 1996;12:515-518.
4. Allavena C, Dousset B, May T, et al. Relationship of trace element, immunological markers, and HIV1 infection progression. Biol Trace Elem Res. 1995;47:133-138.
5. Ripa S, Ripa R. Zinc and immune function. Minerva Med. 1995;86:315-318.
6. Lambl BB, Federman M, Pleskow D, et al. Malabsorption and wasting in AIDS patients with microsporidia and pathogen-negative diarrhea. AIDS. 1996;10:739-744.
7. Sazawal S, Jalla S, Mazumder S, et al. Effect of zinc supplementation on cell-mediated immunity and lymphocyte subsets in preschool children. Indian Pediatr. 1997;34:589-597.
8. Walker CF, Black RE. Zinc and the risk for infectious disease. Annu Rev Nutr. 2004;24:255-275.
9. Brooks WA, Santosham M, Naheed A, et al. Effect of weekly zinc supplements on incidence of pneumonia and diarrhoea in children younger than 2 years in an urban, low-income population in Bangladesh: a randomized controlled trial. Lancet. 2005;366:995-1004.
10. Bobat R, Coovadia H, Stephen C, et al. Safety and efficacy of zinc supplementation for children with HIV-1 infection in South Africa: a randomized double-blind placebo-controlled trial. Lancet. 2006;366:1862-1867.
11. Baum M, Cassetti L, Bonvehi P, et al. Inadequate dietary intake and altered nutrition status in early HIV-1 infection. Nutrition. 1994;10:16-20.
12. Cousins R, Leinart A. Tissue-specific regulation of zinc metabolism and metallothionein genes by interleukin 1. FASEB J. 1988;2:2884-2890.
13. Wieringa FT, Dijkhuizen MA, West CE, et al. Estimation of the effect of the acute phase response on indicators of micronutrient status in Indonesian infants. J Nutr. 2002;132:3061-3066.
14. Duggan C, MacLeod WB, Krebs NF, et al. Plasma zinc concentrations are depressed during the acute phase response in children with falciparum malaria. J Nutr. 2005;135:802-807.
15. Koyanagi A, Kuffo D, Gresely L, et al. Relationships between serum concentrations of C-reactive protein and micronutrients, in patients with tuberculosis. Ann Trop Med Parasitol. 2004;98:391-399.
16. Carcamo C, Hooton T, Wener MH, et al. Etiologies and manifestations of persistent diarrhea in adults with HIV-1 infection: a case-control study in Lima, Peru. J Infect Dis. 2005;191:11-19.
17. Mertz H, Beck C, Dixon W, et al. Validation of a new measure of diarrhea. Dig Dis Sci. 1995;40:1873-1882.
18. Golden MH. The diagnosis of zinc deficiency. In: Sandström B, Mills CF, Eds. Zinc in Human Biology. London, UK: Springer-Verlag; 1989:323-333.
19. Sazawal S, Black R, Bhan M, et al. Zinc supplementation in young children with acute diarrhea in India. N Engl J Med. 1995;333:839-844.
20. Kelly P, Musonda R, Kafwembe E, et al. Micronutrient supplementation in the AIDS diarrhoea-wasting syndrome in Zambia: a randomized controlled trial. AIDS. 1999;13:495-500.

AIDS; diarrhea; HIV; trial of zinc supplementation

© 2006 Lippincott Williams & Wilkins, Inc.