Diarrhea causes an estimated 2.5 million child deaths in developing countries each year, 35% of which are due to acute diarrhea [1]. The importance of zinc and copper in the pathophysiology of acute diarrhea is highlighted by the significantly higher daily fecal losses of these elements during acute diarrheal episodes than the fecal losses in unaffected children [2]. Conversely, zinc supplementation exhibits therapeutic action by facilitating the transport of water and electrolytes across the intestinal mucosa, preventing villous atrophy and improving overall immunity [3]. A pooled analysis of the impact of zinc supplements added to the treatment regimens for acute diarrhea has reported a beneficial effect of reduction in the duration of diarrhea [4]. A recently published Cochrane review has reported a reduction of the duration of acute diarrhea in the zinc group by 12 hours, but there was significant heterogeneity between the trials examined [4, 5]. Consequently, the current paradigm favors the use of Zn supplementation for treatment of acute diarrhea in children [6].

The reported efficacy of zinc therapy for acute diarrhea, although encouraging, needs to be considered in the light of several caveats. First, the initial pooled analysis reported a beneficial effect of zinc on duration of diarrhea but a similar effect on the volume of stools was not included in this analysis or the subsequent Cochrane review [4]. Second, additional supplementations in the form of multivitamins or vitamin A were also administered with zinc and placebo in most of the studies included in these reviews, and it is difficult to ignore the possibility of synergistic effects of the co-interventions. Third, the potential beneficial effects of zinc supplementation across different sub-groups of children with acute diarrhea based on age, nutritional status or diarrheal etiology are still not established [7, 8]. Lastly, community-based trials are more prone to difficulties with monitoring, compliance and measurement of the trial outcomes such as stool frequency, volume and duration. Large community-based, cluster-randomized, double-masked, placebo-controlled trials of daily prophylactic supplements of 10 mg of zinc found no significant difference in frequency of diarrhea, duration, all-cause hospitalization rates or overall mortality, in Nepal, India and Zanzibar [9–11].

It has been speculated that simultaneous depletion of the copper stores can affect diarrhea morbidity and mortality [12], but the preventive or therapeutic role of copper supplementation is still unclear. It is possible that children supplemented with both zinc and copper experience greater reductions in the duration of diarrhea. We therefore conducted a hospital-based double-blind randomized controlled trial to evaluate the efficacy of zinc and copper supplementation when given with standard treatment to children with acute watery or bloody diarrhea. The primary hypothesis of the study was that the children supplemented with zinc or zinc and copper would experience a reduction in the duration of diarrhea, the volume of stool output and the rates of complications, as compared with those receiving placebo.

This was a double-blind, randomized, placebo-controlled clinical trial. The eligibility criteria for this trial were all children aged 6 months to 59 months attending the Indira Gandhi Government Medical College and Hospital in Nagpur, India, with more than three unformed stools in the prior 24 hours; duration of diarrhea up to 72 hours; and ability to accept oral fluids or feeds. Children who were severely dehydrated as per the World Health Organization (WHO) criteria and unable to accept treatment orally were rehydrated with Ringer's lactate and reviewed 4 hours later for ability to take orally [13]. The children were screened for eligibility by a trained study physician. The exclusion criteria were: chronic or severe complicating illness, known positive HIV status, kwashiorkor, residing outside a radius of 30 km around the hospital, participating in another study or already enrolled in this study. The consent procedure was administered to parents or guardians of the children and those who gave informed consent were enrolled and randomized. The Ethics Committee of Indira Gandhi Government Medical College, Nagpur, and the Human Research Ethics Committee of the University of Newcastle, New South Wales, Australia (HREC Approval No: H-500-0203) approved the study protocol, and the treatment effects monitoring committee monitored the trial for safety. The trial is registered with International Standard Randomized Controlled Trial with the unique identifier ISRCTN85071383.

The duration of diarrhea and the total stool output during an episode of acute diarrhea were not significantly different in the supplemented and placebo groups. Other outcome measures such as the use of ORS, IV fluids, episodes of dehydration experienced in the hospital, rates of complications and diarrheal duration more than 7 days were also no different in the children who received Zn or Zn+Cu supplementation as compared with the placebo, when adjusted for nutritional status, type of stool (presence or absence of dysentery) and baseline serum zinc or copper levels. All children received similar doses, frequency and duration of syrups of identical appearance for 14 days and the treatment adherence was similar in the three groups. Serious adverse events related to the syrups were not experienced by any child, suggesting that they can be used safely in treatment protocols.

There could be many reasons for lack of effect observed in our study. We first considered the possibility of either low dose of zinc (an average intake per day of 13.9 mg over 14 days in any of the zinc groups) as compared with other therapeutic studies that have used a fixed dose of 20 mg, or poor treatment adherence, or failure of supplements to replenish the zinc loss, as possible causes of the failure of a favorable response to treatment. However, studies that report administration of fixed dose of 20 mg have not reported the actual mean consumption. Also, in our study there was no difference in treatment adherence across the three groups in the hospital stay or at home after discharge (Table 2). The mean of absolute difference of serum zinc from baseline to the 14th day increased significantly in zinc-supplemented groups, indicating appropriate dose and bioavailability of the supplements (Table 3).

It has been observed that there is a differential extent of zinc deficiency, considered usually as serum zinc less than 60 μg/dl, across study populations, which could explain the variation in effects of either prophylactic or therapeutic zinc supplementation acute diarrhea [7]. It is expected that those countries at high risk of zinc deficiency, that is, prevalence stunting exceeding 20% and estimated prevalence of inadequate zinc intake of more than 25%, would most likely benefit from prophylactic and therapeutic zinc supplementation [18]. However, there is heterogeneity in the effect of zinc during acute diarrhea even in those with low plasma or serum zinc levels. Large studies from Nepal, Zanzibar and India of prophylactic zinc supplementation found no difference in morbidity and mortality of children aged 1 to 48 months between placebo and supplemented groups [10, 11, 19]. However, a smaller prophylactic study from Bangladesh did observe reduced mortality in children, although the majority of the deaths were pneumonia related [20]. The mean baseline plasma zinc in these studies ranged from 62 to 78.7 μg/dl. A therapeutic study from Bangladesh in infants also reported no effect of zinc supplementation and the mean baseline zinc of the study population was 68.5 μg/dl (see [7]).

Also, not all therapeutic studies that reported a beneficial effect of zinc in acute diarrhea were in children with zinc deficiency. Their serum or plasma zinc ranged from 58.0 to 92.9 μg/dl at baseline, with range of percentage change at the end of the therapy being -1.7 to 42, thus failing to establish a clear relation of response to zinc therapy and underlying low blood zinc levels [10, 19, 21–26]. By comparison, in our study the mean baseline serum zinc level was 71.2 μg/dl, with the highest proportion of children with serum zinc at least 60 μg/dl in the Pl group. Although the study outcomes were adjusted for baseline serum zinc and copper, it is possible that there was a risk of longer duration of diarrhea in the intervention groups as they had lower baseline zinc. This may perhaps reduce the observed difference in the diarrheal duration between intervention and placebo groups. However, the overall underweight rate in the study population (weight for age less than -2 Z -score) was 52.4% and was balanced across the groups. A limitation of this study, common to many existing studies of therapeutic and prophylactic zinc supplementation, is that serum zinc concentrations are not a reliable measure of body zinc status [27]. Instead, measurement of dietary zinc or copper intake and tissue zinc or copper status could perhaps explain the differential impact of zinc with respect to varying tissue zinc status.

Age can modify the beneficial effect of Zn supplementation. Two studies with infants younger than 6 months, included in the Cochrane review, showed no impact of zinc on duration of acute diarrhea. Although tissue zinc status was not measured in these studies, they speculated that perhaps the lack of effect was due to adequate zinc stores acquired in utero , through breast feeding and lack of preceding zinc depleting illness [7, 28]. However, five studies that included children aged 2 to 36 months (mean age was 14.4 months), a large proportion of which were also ever breast fed reported a beneficial effect [21, 22, 29–31]. Thus, there is inconsistency in the evidence on whether zinc stores in utero or those acquired from breast feeding would diminish the impact of supplementation on duration of acute diarrhea. In our study, 41% of the children were between 6 and 12 months of age, the mean age was 17.9 months and 56% had received any breast feeding.

The impact of zinc supplementation may also be related to different diarrheal etiology at different age groups and study populations. Zinc has been postulated to have less impact on rotavirus diarrhea and the secretory effect of Escherichia coli heat-stable enterotoxin but beneficial effects on enteropathogenic E. coli infection in animal models [32–35]. The heterogeneity in effects of zinc for acute childhood diarrhea could also be due to a varied patient population, study designs (field based or hospital based), the methods of monitoring diarrhea and the outcome measured. In a pooled analysis of three data sets, the Zinc Investigators Collaborative Group in 1999 [36] reported a multivariate RH of 0.85 (0.78, 0.92) or a 15% lower probability for continuation of acute diarrhea in the zinc group in studies from Indonesia (n = 1,368), India (n = 931), and Bangladesh (n = 101). Although this contrasts with our study, two of the three studies (Indonesia and Bangladesh) showed no difference with RH (95% CI) for continuation of diarrhea of 0.92 (0.83, 1.02) and 0.85 (0.57, 1.28), respectively. The pre-enrolment diarrheal duration for Indonesia, India and Bangladesh was 1.9, 2.7, and 3.4 days respectively, and the mean post-enrolment diarrheal duration was 3.5, 5.1 and 4.5 days, respectively. Only the Indian study showed a mean difference of 1 day from placebo with an effect size of 0.24 (95% CI 0.10, 0.37) in the duration of diarrhea. Further, the Indian and Indonesian studies were community-based trials where cessation of diarrhea was defined as a 48-hour period free of diarrhea (three to four unformed stools in one day) measured from a 5 to 7 day recall of the mother. The Bangladesh trial was a hospital-based study which recruited children with diarrhea for less than 72 hours, monitored 8-hourly in the hospitals for stool output and cessation of diarrhea defined as passage of a soft formed stool. These different study designs indicate that separate meta-analyses for hospital-monitored and community-based studies are required to fully assess the therapeutic effects of zinc in acute diarrhea.

Finally, in several studies it is difficult to ignore the interactions of zinc with multivitamins or co-interventions administered to the study populations. The Zinc Investigators Collaborative Group further conducted a meta-analysis of three of the above mentioned studies and two additional studies [25, 37] from India and Bangladesh. Since multivitamins including vitamin A were administered with zinc in three studies, a potentially beneficial interaction of zinc with vitamin A or other vitamins cannot be ruled out. Zinc is involved in the release of vitamin A from liver cells and in the synthesis of retinol binding protein, and the effect of zinc in infectious diseases may be dependent on an adequate vitamin A status [38–40]. Moreover, there are no studies directly examining the relationship between the different B vitamins and diarrheal disease, which makes it difficult to arrive at any conclusions about their effects on this outcome [30]. It is also noteworthy that blinding is a challenge when combinations of drugs are used. In one of the Indian studies included in a recent meta-analysis, glucose water was used as a placebo which is neither identical in appearance or taste to the intervention, making the double blinding questionable [25].

This trial is the first to evaluate the impact of oral zinc and copper administration on duration of acute diarrhea, serum zinc and copper levels after 14 days of supplementation. Large oral doses and a relatively low dose of dietary zinc are known to interfere with copper bioavailability [41], and contrarily, copper may interfere with zinc absorption and obliterate its effect [42]. A diarrheal morbidity trial of 4 months zinc supplementation in north Indian children reported a significantly lower copper level in the zinc supplemented group, and an adverse effect on copper levels at the end of the study [43]. However, we supplemented zinc and copper in the same ratio as in a customary diet and therefore this Zn-Cu interaction would be unlikely [44, 45]. Also, a significant fall in serum copper levels from baseline was observed only in the placebo group, which had no effect on diarrheal duration. Similar results have been observed in other studies [11, 20, 21].

The results of our study further highlight the heterogeneity of results of zinc supplementation in acute diarrhea. Thus, it appears that therapeutic Zn supplementation may not have a universal beneficial impact on the duration of acute diarrhea. A re-examination of all available trial results, including this trial, is needed to dissect out the potential contributors to heterogeneity of trial results before Zn can be universally recommended for treatment of acute childhood diarrhea.