Effects of diethylene glycol contamination of pharmaceutical products on unexplained acute kidney injury in children: a systematic review

1. Study identification and selection

We performed a qualitative analysis to observe the effect of DEG contamination on the incidence of unexplained AKI in children guided by the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) [14]. Eligible studies published in English before December 2022 were retrieved from the PubMed, EBSCO, and Web of Science (WoS) databases. We manually searched the reference list of each included study to identify other relevant studies.Table 1 lists the search terms used in the databases. After removing duplicates, the authors (MLA, HAS, SBM, AAT, HAF, and AAK) screened the titles and abstracts and then evaluated the full text, if applicable, according to the inclusion criteria. Two authors (MDP and EAC) discussed the retrieved papers and agreed upon which should be included. If necessary, another author (SRS) was consulted to reach consensus. The corresponding authors of the retrieved studies were contacted for clarification as needed.We used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) guidelines to perform a certainty analysis [15].

Table 1.

Study search strategy and terms

2. Inclusion and exclusion criteria

The inclusion criteria focused on the relationship between DEG contamination and the incidence of AKI in children. Clinical signs or laboratory results were approved as DEG contamination; AKI parameters were confirmed by clinical and laboratory results; every observational study design was applicable; review papers, case studies, and case reports that provided actual data were pertinent; proceedings that offered actual data were eligible; and languages other than English were excluded. A case report includes a single case of poisoning, while a case study includes a few cases. Furthermore, editorials, opinions, and perspectives that did not contain actual information or for which full texts were unavailable were excluded. Children were defined as those aged <18 years. From each included study, we obtained data for children and excluded those for adults.

3. Data coding and quality assessment

Coded and extracted data included author information, incidence year, country, study design, population, contamination sources, outcomes, and summary findings. The authors (MLA, HAS, SBM, AAT, HAF, and AAK) independently conducted the data extraction, discussed the findings, and reached consensus. Another author (SRS) was consulted in cases of doubt and redundancy. The synthesis data provided unique statistical findings, including odds ratios (ORs), relative risks, and confidence intervals (CIs). The corresponding authors were contacted if statistical data were missing from the retrieved papers and the effect size was calculated independently. All extracted and tabulated data were analyzed using Microsoft Excel 2017 and Mendeley and stored in cloud services provided by Universitas Islam Indonesia. The quality assessment was performed using the The ROBINS-E tool (Risk Of Bias In Non-randomized Studies - of Exposures), which encompassed 7 risk of bias domains [16]. Three authors (MLA, MDP, and EAC) independently assessed the risk of bias. All authors reviewed and agreed upon the results. The protocol number for PROSPERO is CRD42023416240.

1. Studies of effect of DEG in unexplained AKI

Two case-control studies reported an outbreak of AKI in children in Haiti [8] and Bangladesh [9]. Both elucidated the history of paracetamol administration and identified DEG in children with unexplained AKI. However, only O’Brien et al. [8] revealed a median 14% (1.2%–19.6%) concentration of DEG from 36 acetaminophen syrup analyses. Other antipyretics [9] and vitamins [8] were contaminated as well, although both studies reported that paracetamol elixir, and syrup were contaminated with DEG. A cross-sectional study by Akuse et al. [10] confirmed an unknown AKI outbreak in Nigeria. Their study elaborated on the history of an ingested mixture of paracetamol (120 mg/5 mL) with diphenhydramine hydrochloride and reported that brand name my pikin (6.25 mg/5 mL) had been contaminated with DEG. Of the 95 children with AKI who consumed several drugs prior to being hospitalized, 60 ingested my pikin syrup, 56 consumed other brands of paracetamol, 23 used herbal medicines, and 6 consumed other brands of teething tonics [10].

Five case studies reported DEG in an epidemic of AKI of unknown etiology in Nigeria [17,18], India [12], and South Africa [13] in which total of 47 children were admitted to the hospital [17]. Obebi Cliff-Eribo et al. [18] in Nigeria reported that 7 cases of AKI were correlated with the use of contaminated paracetamol/diphenhydramine chloride and herbal medicines for teething problems. This study revealed that 4 of 7 patients ingested paracetamol and the remaining patients consumed herbal medicine. Singh et al. [12] reported that 17.5% (v/v) of unknown AKI outbreaks in India were caused by DEG-contaminated cough syrups (4.25-mg bromhexine hydrochloride and 27.4-mg pseudoephedrine hydrochloride per 5 mL). Similar to the report by Singh et al. [12] in India, Hari et al. [19] reported that 11 children suffered from unexplained AKI due to the use of DEG-contaminated paracetamol, with a 2.3%–23% w/w concentration. Additionally, 7 children in South Africa were diagnosed with unexplained AKI caused by DEG-contaminated sedative use [13].

Most studies have been case reports of a single patient with DEG-induced unexplained AKI [7,20-29]. Some case reports described unintentional EG contamination or DEG-induced oliguria or anuria [21,24,27,29] after the consumption of a DEG-contaminated acetaminophen elixir [26], PG-containing intravenous phenobarbital and pentobarbital for uncontrolled seizures [28], DEG ingestion suicide attempts [7], ergocalciferol-containing high-dose vitamin D dissolved in PG for diagnosing nutritional rickets [22], intentional EG poisoning in baby formula [25], DEG-contaminated fever and cough medication [23], and EG contamination of laxatives [20].

High-quality studies of AKI induced by DEG and other glycol-induced AKI are limited. Our study summarized that most studies revealed that the outbreak of unexplained AKI originated from DEG-contaminated pharmaceuticals, including paracetamol, a mixture of paracetamol and diphenhydramine hydrochloride, a mixture of bromhexine HCL and pseudoephedrine HCL, a sedative mixture, adverse drug reactions from the teething tincture, unspecific cough syrups, and unintentional poisoning. Few studies have reported intentional or unintentional EG poisoning and PG-containing high-dose drugs.

2. DEG-contaminated paracetamol-induced unexplained AKI

DEG-contaminated paracetamol has primarily been reported. Any DEG-contaminated paracetamol (OR, 18.2; 95% CI, 4.4–75.2; P<0.001) [9] and syrup (OR, 52.7; 95% CI, 15.2–197.2) [8,17] was strongly correlated with unexplained AKI in children. Among 339 children with unknown AKI in the study of Hanif et al. [9], paracetamol elixir ingestion was 13.3 times more likely to cause unexplained AKI compared to 90 cases of identified AKI, any fever medication among 308 children was 22 (12.2–39.3; P<0.001) times more likely to cause unexplained renal failure paralleled to 28 cases of known AKI; the ingestion of DEG-containing branded paracetamol was confirmed among only children with unexplained AKI [9]. That study also revealed that 19 of 69 paracetamol elixirs from different brand products contained DEG as a diluent, while the remaining 50 paracetamol elixirs contained PG as a diluent. However, this study could not obtain elixir samples for further analysis from patients that may have been strongly contaminated with DEG. Therefore, the unexplained AKI outbreak in Bangladesh caused 185 deaths due to AKI and 157 deaths (85%) due to the consumption of an unknown elixir for fever [9].

A similar study by Hari et al. [19] reported that 11 children suffered unexplained AKI, 6 of whom had a history of ingesting elixir paracetamol containing 2.3%–22.3% w/w (median: 15.4 w/w). Scalzo [26] corroborated former studies reporting that contaminated elixir paracetamol caused unexplained AKI. However, this case study revealed no DEG or parent compounds in the patient specimen analysis. DEG-contaminated paracetamol was confirmed based on the history of ingestion in outbreak cases in Haiti.

O’Brien et al. [8] investigated 109 cases of unexplained renal failure caused by contaminated paracetamol syrup. Paracetamol syrup consumption revealed an OR of 52.7 (95% CI, 15.2–197.2) compared to other drugs, and the remaining drug evaluation confirmed no statistical significance on unexplained AKI, although a laboratory examination revealed that these drugs were also contaminated with DEG. Further statistical analyses were performed when DEG-contaminated drugs demonstrated an OR of 44.2 (95% CI, 13.4–156.6) [8]. The median DEG concentration of 36 bottles was 14.4% (95% CI, 1.2%–19.6%); this was the equivalent of 1.34 mL/kg in 32 children (95% CI, 0.22–4.22 mL/kg) and 1.0 mL/kg in 12 children. The primary source of DEG contamination in this study was a formulation with 50% glycerin, which was comparable to the 12% DEG concentration in the final products [8].

Okuonghae et al. [17] reported that 47 pediatric patients died after ingesting paracetamol syrup. However, this study revealed that paracetamol syrups were not the only drugs consumed by the patients; other drugs such as chloroquine, promethazine, and cough syrups were also administered to them. Laboratory examinations revealed no information about the concentrations of these drugs. Thus, predicting paracetamol as a primary source of contamination is impossible because a few drugs were ingested.

Akuse et al. [10] revealed that the ingestion of my pikin teething syrups composed of paracetamol (120 mg/5 mL) and diphenhydramine hydrochloride (6.25/5 mL) caused unexplained acute renal failure in 60 of 119 children. Of the 119 children with unknown AKI, 95 consumed drugs before developing the disease, including my pikin by 60, other paracetamol brands by 56, native herbs by 23, and other brands by 6. However, only my pikin ingestion showed statistically significant differences compared to the other drugs. Furthermore, laboratory evaluations of 6 of the 7 residue bottles confirmed 17%–21% DEG contamination by weight in the sample [10]. This study could not reflect the actual exposure to DEG in my pikin because the children also ingested other drugs and the DEG concentration was not examined. The examination of 6 residue bottles was unable to conclude that my pikin was the primary source of the poisoning, as other brands of paracetamol existed.

Obebi Cliff-Eribo et al. [18] reported a similar series of 8 patients with unexplained drug-induced AKI. The authors concluded that 4 cases were caused by DEG in my pikin teething mixture, 3 by the herbal teething mixture, and one by dihydroartemisinin. Additionally, diphenhydramine is not the only antihistamine that causes AKI; promethazine has also contributed. All 56 cases of unknown AKI in children in Gambia revealed that a mixture of paracetamol and promethazine syrup was the cause [11].

3. Contribution of other contaminated drugs to unexplained AKI

DEG-contaminated paracetamol is not the only cause of vehicle poisoning. Other drugs such as cough expectorants [12,23], sulfanilamides [30], and sedatives can contribute [13]. Apart from the history of paracetamol contamination, Singh et al. [1] revealed interesting findings that a cough reliever was a primary source of unknown AKI in 36 children. However, their investigation revealed that only one cough syrup, comprising 4.25 mg of bromhexine hydrochloride and 27.4 mg of pseudoephedrine hydrochloride, was contaminated with DEG. Furthermore, one nonspecific cough syrup sample was contaminated with EG, one paracetamol syrup sample was negative for any contamination, and 4 drugs, including paracetamol and cough syrup from different manufacturers, were negative for DEG and EG [12]. One sample in the cough syrup investigation was insufficient to conclude that the drug was vehicle-contaminated. Additionally, the paracetamol drugs that were reportedly contaminated tested negative for harmful substances. Therefore, this study does not address the relationship between exposure and outcomes. Jain et al. [23] confirmed that cough syrup use was associated with unexplained AKI in a boy with fever and cough. This study did not reveal the drug’s ingredients; however, we assumed that the cough syrup contained paracetamol and expectorants.

One study recognized sedative agents as the cause of death of 7 children [13]. They investigated that 2 brand name sedatives (Pronap and Plaxim) substituted PG for DEG during the manufacturing process. All sedative drugs used in their study were obtained from parent interviews. However, the data were limited due to a lacking further examination of DEG concentrations and statistical analysis, making this study prone to bias.

PG overconcentration and EG contamination are common medications, such as antiseizure agents [28], vitamin D treatments [22], and laxative agents [20]. One study reported the adverse drug effects of high-dose intravenous phenobarbital and pentobarbital for uncontrolled seizures after the failure to prescribe standard medicine for seizures. During the 8-day treatment, 6.53 g/day of pentobarbital and 6.70 g/day of phenobarbital were infused, comparable to the ingestion of 90.3 g/day of PG.Another study revealed that oral vitamin D contains 100,000 international units (IU) of ergocalciferol every 2 hours for a total of 600,000 IU over 12 hours, which induces unwanted AKI [22]. Ergocalciferol contains 103.6 g/100 mL of PG, and 600,000 IU is equivalent to 75 mL of the solution or 77.7 g of PG. Another study by Aly et al. [20] revealed that an EG-contaminated laxative was the primary cause of AKIin a boy with a complicated congenital disease. Based on the renal biopsy evaluation, the drugs were suspected to be contaminated with EG. Several studies of PG and EG have been performed on a single patient without a standardized methodology. The patients had nutritional and congenital comorbidities that probably affected their homeostatic functions during treatment. Thus, knowledge of the adverse effects of PG during treatment is limited due to its common use as a vehicle to enhance drug solubility.

4. Other factors related to unexplained AKI

Several cases of unintentional DEG toxicity and parent compound-induced unwanted AKI have been confirmed, and some were intentionally poisoned. Vale et al. [29] reported the case of a child who accidentally ingested 100 mL of EG, comparable to 26.9 g of EG and 73 mg of oxalate, on dialysis. Similarly, Walker et al. [27] fortuitously ingested nonspecific drugs suspected of containing EG. They postulated that the poisoning caused kidney enlargement, and a biopsy revealed crystal oxalate. Another study by Harry et al. [21] confirmed unwanted ingested antifreeze contained 41% of EG. Unwanted AKI was diagnosed based on laboratory examinations, such as metabolic acidosis with an anion gap of 29 mmol/L, an osmolar gap of 50 mOsm/L, and a chromatography result of 3.1 g/L. The last unintentional case was reported by Brophy et al. [24], who revealed accidental poisoning combined with DEG using tetraethylene glycol and penta-ethylene glycol. They reported no metabolic acidosis or anion and osmolar gaps; otherwise, chromatography revealed 1.7-mg/dL DEG. Most cases of unintentional poisoning have been reported in single patients without robust scientific analyses and statistical methodologies. Moreover, one study did not report the EG concentration in the specimens, resulting in bias of the reporting results.

Two other intentional cases of DEG poisoning were suicide [7] and assassination [25]. Gupta et al. [7] revealed that DEG was used in suicide attempts by the ingestion of brake fluid containing DEG as the primary constituent. A laboratory evaluation revealed an osmolar gap, and chromatography revealed a DEG concentration <1 mg/dL. Greenky et al. [25] described a case of the intention to kill an 8-day-old neonate by the addition of DEG to milk formula. The laboratory findings indicated signs of poisoning, including metabolic acidosis, anion and osmolar gaps, and the presence of oxalate. However, the DEG concentration was not determined. Two described cases of unintentional and intentional poisoning included limited data. Thus, we could not reflect on the exposure and outcome assessments in this case. Furthermore, this finding indicates that DEG and EG are commonly neglected owing to their accessibility and affordability in the market.