Review Article

The efficacy of hemodialysis on paraquat

poisoning mortality: A systematic review and
meta‑analysis
Nastaran Eizadi‑Mood1, Danial Jaberi2, Zahra Barouti2, Alireza Rahimi3, Marjan Mansourian4, Gholamali Dorooshi5,
Ali Mohammad Sabzghabaee5, Sam Alfred6
1
Department of Clinical Toxicology, Isfahan Clinical Toxicology Research Center, Isfahan University of Medical Sciences, Isfahan, Iran,
2
School of Medicine, Isfahan Medical School, Isfahan University of Medical Sciences, Isfahan, Iran, 3Clinical Informationist Research Group,
Health Information Technology Research Center, Faculty of Medical Management and Information Sciences, Isfahan University, Medical
Sciences, Isfahan, Iran, 4Department of Epidemiology and Biostatistics, Isfahan University of Medical Sciences, Isfahan, Iran, 5Isfahan Clinical
Toxicology Research Center, Isfahan University of Medical Sciences, Isfahan, Iran, 6Department of Emergency Medicine, Royal Adelaide
Hospital, University of Adelaide, South Australia

Background: Paraquat (PQ) poisoning is a serious public health concern, especially in developing countries, due to its easy access
and lack of awareness of potential harms. No effective treatment has been reported yet. Conventional hemodialysis (HD) is still
used in many centers for excreting PQ or reducing acute kidney injury, but there is no consensus on its efficacy. Therefore, we
aimed to review the HD efficacy in PQ poisoning mortality. Materials and Methods: We searched Web of Science, PubMed,
Excerpta Medical Database, Google Scholar, Scopus, Cochrane, Web of Knowledge, Pro‑Quest, ScienceDirect, Springer, Clinical
Key, Scientific Information Database, Magiran, and Iran‑doc, in publications before January 1, 2020. We compared patients
who underwent HD (Group 1) with those who did not (Group 2). The outcome was considered mortality/survival. The data
were analyzed by Comprehensive Meta‑analysis Software. Results: This systematic review and meta‑analysis included five
studies with a combined total of 203 patients. The patients in the Group 1 had higher mortality than Group 2 (odds ratio, 2.84;
95% confidence interval: 1.22–6.64; P = 0.02). There was no evidence of publication bias (P value for Egger’s test = 0.833).
Conclusion: Although HD did not affect the survival of patients, other variables such as the amount of ingested PQ, poisoning
severity, the time between PQ ingestion and the start of HD, duration, and times of HD sessions may influence the results
regarding mortality.

Key words: Meta‑analysis, mortality, paraquat, poisoning, survival, systematic review

How to cite this article: Eizadi‑Mood N, Jaberi D, Barouti Z, Rahimi A, Mansourian M, Dorooshi G, et al. The efficacy of hemodialysis on paraquat
poisoning mortality: A systematic review and meta-analysis. J Res Med Sci 2022;27:74.

INTRODUCTION occurs in severe poisoning. The most common cause of

death is respiratory failure.[2‑5]
Paraquat (PQ) is an effective herbicide with a high
mortality rate in acute poisoning.[1,2] Exposure occurs There is still no standard and effective treatment.
accidentally or intentionally as a suicide attempt. PQ
poisoning is becoming a serious public health concern, Decreased absorption by gastric lavage and oral
especially in developing countries, due to its easy access activated charcoal; elimination through different
and lack of awareness of potential harms. Multiple techniques (conventional hemodialysis [HD],
organ failure, including lung, kidney, and liver failure, hemoperfusion [HP], continuous venovenous
hemofiltration [CVVH], and continuous renal
Access this article online
This is an open access journal, and articles are
Quick Response Code:
Website: distributed under the terms of the Creative Commons
www.jmsjournal.net Attribution‑NonCommercial‑ShareAlike 4.0 License, which
allows others to remix, tweak, and build upon the work
non‑commercially, as long as appropriate credit is given and
DOI:
the new creations are licensed under the identical terms.
10.4103/jrms.jrms_235_21
For reprints contact: WKHLRPMedknow_reprints@wolterskluwer.com

Address for correspondence: Dr. Alireza Rahimi, Medical Informatics Department, Health Information Research Center, Isfahan University Medical
Sciences, Isfahan, Iran.
E‑mail: a_rahimi@mng.mui.ac.ir
Submitted: 11‑Mar‑2021; Revised: 22‑Feb‑2022; Accepted: 03‑Mar‑2022; Published: 27-Sep-2022

1 © 2022 Journal of Research in Medical Sciences | Published by Wolters Kluwer - Medknow | 2022 |
 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

replacement therapy [CRRT]); decreased inflammatory without other treatment modalities such as gastrointestinal
response by the immunosuppressants (corticosteroids and decontamination, immunosuppressant (dexamethasone,
cyclophosphamide); antioxidants (N‑acetyl cysteine [NAC], methylprednisolone, and cyclophosphamide), and
Vitamin C, and Vitamin E); and a combination of different antioxidants (NAC, Vitamin C, Vitamin E, and
therapies have been used to treat patients with PQ deferoxamine). C (comparison): Comparison between
intoxication with different outcomes for survival.[6‑29] In patients who received HD (Group 1) with those who did
addition, prognostic factors have been reported in some not (Group 2). The other standard treatments in both groups
studies concerning different treatments.[30,31] were similar. O (outcome): Assessment of death or survival.
Three researchers evaluated the full texts of included
Conventional HD is used in many developing countries for studies independently. Finally, in a consensus meeting,
patients with PQ intoxication. Some studies have shown that the research team, according to inclusion/exclusion criteria,
HD reduces mortality, while others have reported that HD decided whether the study should be accepted for quality
has little effect on patient recovery.[4,9,10,29,32‑34] and quantity analysis or not.

A l t h o u g h t h e e f f e c t o f H P , C V V H , C R R T, We did not apply any limitation on article types and

immunosuppressants, and antioxidants has been reported searched all interventional (randomized clinical trial),
in PQ poisoning in systematic review studies,[35‑43] the impact noninterventional, and observational studies that included
of HD on the outcome of patients with PQ poisoning has all historical cohorts, cross‑sectional, and case–control.
not been evaluated systematically yet. Due to the high Furthermore, one researcher reviewed all articles’ references
rate of PQ patients, a systematic review and meta‑analysis to find more relevant studies in the second round of the
study was conducted on the HD effect on the outcome of project (hand searching).
the patients (survival/death).
The inclusion criteria were: (1) articles published in English/
MATERIALS AND METHODS non‑English and Persian (with English abstract) before
January 1, 2020; (2) the studied population was PQ poisoning
This study was a systematic review, and a meta‑analysis patients in all age ranges; (3) some patients were treated
was carried out according to the PRISMA guideline.[44] The with HD (Group 1), and others were not (Group 2); (4) other
Institutional Ethics Committee of Isfahan University of treatment modalities such as immunosuppressants and
Medical Sciences approved the project. antioxidants were similar in both groups; (5) the outcome
had been reported as death or survival; and (6) the full text
Search strategy and study selection of the articles was available. In the case of articles whose full
Three specialists of the research team searched the text was not available, the necessary data could be obtained
international reference databases: Web of Science, PubMed, from the summary of the article we included. Otherwise, we
Excerpta Medical Database, Google Scholar, Scopus, e‑mailed the corresponding author to access the full text.
Cochrane, Web of knowledge, Pro‑Quest, Science Direct, If we did not receive a response, that article was excluded.
Springer, Clinical Key, Scientific Information Database, To find gray sources, we performed a particular search in
Magiran, Iran‑doc, Ministry of Health (MOH) Thesis, and related databases such as clinical trial records (e.g., http://
MOH articles, in publications before January 1, 2020. We www.irct.ir/, http://www.trialscentral.com/, https://www.
selected all studies with the following keywords: Paraquat, proquest.com/, and http://www.gateway.com/worldwide/).
hemodialysis, dialysis, poisoning, overdose, intoxication, The exclusion criteria were as follows: (1) all animal studies,
mortality, survival, fatality. Our search strategy was as the case reports, and narrative review articles, and (2) route of
following patterns: ((hemodialysis OR dialysis or “Renal exposure was through the skin, eye, or inhalation.
Dialysis” OR hemodialysis) AND (paraquat OR Gramoxone)
AND (poisoning OR intoxication OR overdose OR toxicity) Data extraction
AND (survival OR death OR mortality OR fatality)) in Two researchers independently extracted information from
TITLE, as well as ((haemodialysis OR dialysis or “Renal each article and recorded them in an excel table. These data
Dialysis” OR hemodialysis) AND (paraquat OR Gramoxone) included bibliographic details (name of the first author, year
AND (poisoning OR intoxication OR overdose OR toxicity) of publication, country, type of study), participant’s number
AND (survival OR death OR mortality OR fatality)) in in each group, characteristics (gender, age), PQ‑ingested
KEYWORD/MESH/SUBJECT [Supplementary Table 1]. dose, co‑ingestion of PQ with other substances, performing
Three researchers independently conducted the screening sodium dithionite test, urine or blood PQ level, time
process by PICO search tool: P (population): Patients with from ingestion of PQ to admission in the hospital, time to
PQ poisoning either accidentally or intentionally as a performing HD, duration and number of HD times, other
suicide attempt. I (interventions): Performing HD with or treatment modalities, presence or absence of underlying

| 2022 | Journal of Research in Medical Sciences 2

 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

disease, mortality/survival, and study ethical approval/or Statistical analysis

trial registration. The observed differences were evaluated We pooled the relative association measures of outcome
by the third researcher in a consensus meeting. from included studies. We analyzed the data using the
fixed‑effects model if there was homogeneity and the random
Quality assessment fixed‑effect model if there was heterogeneity described by
We evaluated the quality of the included studies using the DerSimonian and Laird.[48] Heterogeneity in results across
Joanna Bridges Institute (JBI) for Integrated Information studies was determined using the I2 statistic, which assesses
Management, Evaluation, and Review system. [45,46] the proportion of variation in results across studies. An I2 of
Any approved article is eligible for design as an expert 50% or more indicates a considerable inconsistency between
consensus. The first quality assessment tool used was the JBI studies. Meta‑analysis was completed using Comprehensive
checklist for cross‑sectional analytical studies. This checklist Meta‑analysis version 2.2 (Biostat Inc., Englewood, NJ,
was applied to the four cross‑sectional studies in this study. USA). Publication bias was estimated using Egger’s
The checklist included eight items that enable the critical regression test. In this regression, the treatment effect size
evaluation of studies for possible biases. The second JBI and bias are captured by the slope of the regression line and
tool used was the cross‑sectional checklist for two specific the intercept, respectively.[49] The odds ratio (OR) and 95%
studies in our review and two cross‑sectional studies with confidence interval (CI) were estimated and included in a
a series of four and six patients. This checklist includes forest plot. The significance level that defined the existence
ten items for evaluating these studies’ quality assessment of study heterogeneity was set as <0.05.
in the following aspects: erosion selection, performance,
diagnosis, and bias, as well as pilot designs. In addition, Dealing with unclearly presented data
the randomized trial JBI checklist assessed the following We sent an e‑mail to the corresponding/first authors of
items for one RCT in our review; selection, performance, articles regarding unclear information.[4,9,29]
detection, attrition bias, and trial designs. The four answers
indicate the degree to which the article meets the criteria; RESULTS
yes, no, unclear, and not applicable. Finally, these studies
were scored based on Yes cases.[47] Two researchers scored Figure 1 shows study selection (PRISMA flowchart). We
articles by relevant checklist independently. Finally, we could not find any studies that conclusively compared the
resolved the disagreement on the assessment through outcome (survival/mortality) of patients treated with HD
discussion sessions until we reached a consensus. compared with those not treated with HD considering

Figure 1: A flowchart of literature searches in this meta-analysis

3 Journal of Research in Medical Sciences | 2022 |

 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

inclusion criteria. However, among all the studies, we found and more than 100 ml as severe. Three patients had severe
six studies, with some patients being treated with HD and poisoning.
others not receiving HD. Then, we extracted mortality and
survival rates from these articles and considered them for Study # 3
systematic review. A cross‑sectional study performed by Delirrad et al.
on 41 patients suggested that HD had no significant
Studies’ characteristics association with clinical outcomes.[29] In‑hospital mortality
We found one randomized trial study,[34] three cross‑sectional in their center was 46.4%. HD was performed in 92.7% of
studies,[4,9,29] and two cross‑sectional studies consisting of their patients. The amount of PQ ingestion significantly
four and six patients.[10,33] Tables 1 and 2 summarize all affected mortality (149.3 ± 225.4 ml in 36 cases), and it was
studies’ characteristics. Following are some of the results higher in the patients who received HD. HD started in
of individual studies. the 6.73 ± 12.7 h after hospital admission with an average
frequency of 1.46 ± 0.84 times. They found no relationship
Study # 1 between HD, its frequency, starting time, and outcome of
52 adults were assigned to one of the four groups based on the patients.
the risk of mortality (high, borderline, low, or unknown)
in the study of Proudfoot et al.[9] PQ intoxication had Study # 4
been confirmed by measuring plasma PQ concentration Another cross‑sectional study by Kavousi‑Gharbi et al. on
or urine quality test. Their results showed that dialysis 104 PQ‑poisoned patients recommended HD, which was
could not reduce the final concentration of PQ; however, immediately performed on these patients.[4] Eighty‑eight
the mortality of patients in the high‑risk and borderline patients received HD (two patients were discharged after
group who receive HD was lower than that of patients HD with their consent without follow‑up, 44 patients
who do not receive HD. Furthermore, patients with low survived, and 42 patients died). Sixteen patients did not
or marginal toxicity, who had undergone HD, might receive HD (two patients were discharged with their own
recover. We included 35 patients in this study in the consent, 11 patients survived, and three died early before
meta‑analysis and excluded 17 patients (one patient was performing HD). Therefore, we excluded four patients from
treated with peritoneal dialysis, and 16 patients were the meta‑analysis (those discharged with their own consent).
treated with both peritoneal and HD). Twenty‑three took There was a significant difference between outcome and
formulations containing 20% PQ (13 cases in high‑risk, performing HD. The amount of ingested PQ in patients
5 cases in borderline, and four instances in low‑risk was 34.61 ± 55.36 ml. Seventy‑six patients had vomited
groups, and one patient was uncertain of the amount before admission, which may reduce the toxicity. Time to
of ingestion). Twenty‑eight patients drank solutions of admission was not available. Three cases had died due to the
Weedol (2.5% w/w PQ) (3 patients in high, one patient in severity of poisoning during the early hours of admission
borderline, and 24 patients in low‑risk groups). One of the before performing HD; nonetheless, initiating HD was
remaining two ingested a solution containing 8.8% (w/v) delayed in all cases, at least for 6 h, mainly due to the delay
PQ (low), and the concentration in the other was never in receiving the results of viral marker status. Moreover, 54%
discovered. of the patients received HD again due to the increased renal
biomarkers. Based on receiver operating characteristic curve
Study # 2 analysis, they suggested that consuming more than 22.5 ml
Pavan[33] reported six cases of PQ poisoning. Mortality was of 20% PQ can lead to a poor prognosis in the patients.
66% (80% in the HD group). Acute kidney damage occurred
in all cases. Respiratory and multiple organ failure were Study # 5
the leading causes of death. They concluded that dialysis Cook et al.[10] presented serial lung function studies in six
could not remove PQ, and HD is supportive care for kidney cases of PQ intoxication. Two patients were excluded from
failure. Of the six patients, one patient was hospitalized 5 h the meta‑analysis because they were treated with both
after accidentally swallowing a few drops of PQ and did peritoneal and HD methods. Two patients got HD and died.
not receive HD. He survived without HD. The other five The other two patients did not undergo HD; one was treated
patients who arrived at the hospital between 1 and 5 days with forced diuresis and eventually underwent a lung
late received HD due to acute kidney damage. One patient transplant. He was 15 years old when he ingested a 20% PQ
who ingested a mouthful of PQ in the HD group survived and died 19 days later. Another patient was admitted with
with a relative improvement in kidney function. The amount hypoxia and was treated with adjuvant ventilation because
of PQ consumed in other patients was up to 100 ml. In this of the lack of a suitable donor for lung transplantation. He
study, patients were categorized based on the PQ ingestion presented to the hospital 4 days after swallowing and died
amount; less than 20 ml as mild, 20–100 ml as moderate, 26 days later. The result of this study was not shown in

| 2022 | Journal of Research in Medical Sciences 4

5
Table 1: Different variables of included studies in the systematic review
Author/year Study design/ Sample size Group 1 (HD) Group 2 (NHD) PQ blood PQ urine DT Time to HD (h) Duration/ In hospital JBI
area (HD/NHD) levels level times of HD mortality (%) scoreb
Cooke et al., Cross‑sectional 4 (2/2)a 1 case: HD + lung 1 case: Assisted ventilation + (1 case) + NM Different in Different in Group 1: 100% 7/10
1973[10] (case series) transplant 1 case: Con. (forced each patients each patients Group 2: 100%
1 case: HD + Con. diuresis) + lung Minimum:
(forced diuresis) transplantation 3 days
Ponce et al., Randomized 17 (7/10) HD+Con. (GL, Plasmapheresis + Con. NM NM + NM Minimum: 12 Total mortality: 64% 10/13
1986[34] trial/Portugal Fuller’s earth), (GL, Fuller’s earth), h/5 times Group 1: 71%
(Lisbon) methylprednisolone methylprednisolone Group 2: 54%
Proudfoot Cross‑sectional 35 (9/26) HD NM + + + Mentioned Mentioned for Group 1: 32.07% 6/8
et al., 1987[9]η (retrospective)/ for 4 cases: 4 cases Group 2: 33.33%
UK 0.5/2.5/4.3/10 2 cases: 8 h
h 2 cases: 6 h
Mean: 4.32 Mean: 7
Pavan, 2013[33] Cross‑sectional 6 (5/1) 3 cases: HD + Con. (AC, GL, mannitol NM NM + NM NM Group 1: 80% 7/10
(case series) methylprednisolone and normal saline), Group 2: 0%
l/India 1 case: HD + Con. (AC), IV furosemide,
methylprednisolone methylprednisolone
1 case: HD + Con. (AC
GL), furosemide IV,
methylprednisolone
Delirrad et al., Cross‑sectional 41 (38/3) HD + Con. (GL, Con. (GL, AC and sorbitol), NM NM + 6.73±12.7 NM/1.46±0.84 Total mortality: 5/8

Journal of Research in Medical Sciences

2015[29] (retrospective)/ AC, sorbitol), immunosuppression/ times 46.4%
Iran (West immunosuppression/ corticosteroid/NAC Group 1:50%
Azerbaijan) corticosteroid/NAC Group 2: 0%
Kavousi‑Gharbi Cross‑sectional 100 (86/14) HD + Con. (GL, AC, Fuller’s Con. (GL, AC, Fuller’s NM NM NM Minimum: 6 NM Total mortality: 6/8
Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

et al., 2017[4] (retrospective)/ earth), corticosteroid/ earth), corticosteroid/ 43.3%

Iran (Shiraz) cyclophosphamide/NAC, cyclophosphamide/NAC, Group 1: 48.8%
Vitamin E, Vitamin C Vitamin E, Vitamin C Group 2: 21.4%
a
Two cases excluded because they received both peritoneal dialysis and hemodialysis, bMaximum JBI score: Randomized trial=13, cross‑sectional study=8 and case series study=10, ηThe total number of patients in the study was
52. 17 patients received PD or both HD/PD, therefore excluded from analysis. DT=Dithionate test; HD=Hemodialysis; NHD=Not received HD; PQ=Paraquat; NM=Not mentioned; AC=Activated charcoal; Con.=Conventional therapy;
GL=Gastric lavage; NAC=N‑acetyl cysteine; PD=Peritoneal dialysis

| 2022 |
| 2022 |
Table 2: The comparison of different variables in include studies in the systematic review with respect to toxico‑demographic factors
Author/year Age (HD), year Age (NHD), Time to Time to Amount of Amount of Time to Time to HD ARF/ARF start ARF/ARF start
year admission admission PQ (HD) PQ (NHD) death (HD) death indication day (HD) day (NHD)
(HD) (NHD) (NHD)
Cooke et al., 1 case: 18 years 1 case: 28 years 1 case: 3 days 1 case: 4 days 1 case: Mouthful 1 case: 20 g 2 cases: 1 case: 2 cases: 2 cases/4th day 2 cases/12th day
1973[10]^ 1 case: 46 years 1 case: 15 years 1 case: 4 days 1 case: NM 1 case: Mouthful granules PQ 5% 12 days 26 days ARF and 6th day
PQ 20% 1 case: Mouthful 1 case: No requiring
PQ 20% 19 days dialysis
Ponce et al., Mean: 42.5 years (13–72 years)* Death in first 24 h: 6–72 Death in first 24 h: 50–250 NM NM PQ decline None of these patients had ARF*
1986[34] h (mean: 21.2 h) ml (mean: 110 ml)
Death after 24 h: 6–48 Death after 24 h: 15–50
h (mean: 16.5 h) ml (mean: 32.5 ml)
Survived: Mean 21 h* Survived: Mean 52 ml*
Proudfoot NM NM NM NM NMa NM NM PQ decline NM NM
et al., 1987[9]η
Pavan, 2013[33] Mean: 33.6 28 years 1 case: 1 day 1 case: 5 h 1 case: 100 ml A few drops (1 2 cases: Survived 4 cases: 5 cases/NM 1 case/NM
(28–37) 1 case: 2 days 1 case: 50 ml ml) 4 days AKI
2 cases: 3 days 1 case: A large 1 case: 6 days 1 case:
1 case: 5 days cup (250 ml) 1 case: 7 days PQ decline
1 case: A half 1 case:
spoonful (15 ml) Survived
Delirrad et al., Mean±SD: Mean±SD: 28±2 <6 h: 65.8% <6 h: 33.3% 160.9±232.1 ml 21.67±17.56 ml NM NM NM Total ARF: 33.3%
2015[29] 31.9±17.5 (26–28) 6–24 h: 23.7% >24 h: 66.7% 63.2% (1 case)/1st day
(16–75) >24 h: 10.5% ARF at 1st day:
31.6%

Journal of Research in Medical Sciences

ARF at 2nd day:
26.3%
ARF at 3rd day:
2.6%
Kavousi‑Gharbi Female mean age: NM NM Total: 34.61±55.36 ml (1.5–300 ml) 4.8±4.62 days* PQ decline 54% of the Expired due to
Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

et al., 2017[4] 22.81±9.87 years Not survived patients: 66.63±72.61 patients were the severity of
Male mean age: 27.21±11.06 years* ml hemodialyzed poisoning during
Survived patients: 10.18±5.77 ml* again due to the early h of
increase in renal admission and
biomarkers after before HD
the first day
*=Data for HD group and non-HD group not mentioned separately. η=The total number of patients in the study was 52. 17 patients received PD or both HD/PD, therefore excluded from analysis; PD=Peritoneal dialysis; PQ=Paraquat;
SD=Standard deviation. ^=Two cases excluded because they received both peritoneal dialysis and hemodialysis

6
 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

the meta‑analysis because the final result (mortality) of the Second meta‑analysis result
meta‑analysis was the same for both groups. In the second meta‑analysis, we included the data regarding
those studies that HD performed for PQ elimination
Study # 6 before acute kidney injury (AKI). Those studies that
In the study of Ponce et al., [34] 17 patients with PQ patients admitted late and HD was performed for AKI
poisoning were assigned prospectively in two modalities: were excluded. Figure 3 presents the results of the second
HD (7 patients) and plasmapheresis (10 patients). analysis, including three studies. The patients in the
Both groups also received standard pulse therapy and Group 1 had a higher risk of mortality than patients in
methylprednisolone for 3 days. Total mortality was Group 2 (I2 = 2%, P value for heterogeneity = 0.587; Q = 1.067,
64% (71% in the HD group and 54% in the plasmapheresis τ2 = 0.00, and P value for Egger’s test = 0.698).
group). Plasmapheresis showed promising results in this
study. This study was included in the systematic review; The quality assessment checklist of these studies showed that
all of these studies had a good‑quality score (more than 7).
however, we exclude it from the meta‑analysis, as one group
received plasmapheresis.
Potential biases
Publication bias is an issue in every systematic review. We
The results regarding the quality of studies are presented
conducted this review based on the predefined inclusion
in Supplementary Tables 2‑4.
criteria and methodology to select and appraise eligible
studies. The search for studies was extensive and was
First meta‑analysis result
conducted on English and Persian databases. Although
The first meta‑analysis includes five studies with a total of
there were only six studies included in this systematic
203 patients. We did not show the study of Cook et al. in the
review, we believe that due to the extent of the search
meta‑analysis figure because mortality and survival were
considering inclusion and exclusion criteria, these were
the same.[10] The results showed no evidence of publication the only studies addressing this research question at the
bias (I2 = 0%, P value for heterogeneity = 0.386; Q = 1.146, time of the search.
τ2 = 0.00, and P value for Egger’s test = 0.833). The patients
in the Group 1 had a higher risk of mortality than patients DISCUSSION
in Group 2 (OR, 2.84; 95% CI: 1.22–6.64; P = 0.02) [Figure 2].
The tau and Q values are added to the figures. Because the Overall applicability of evidence
heterogeneity was between studies according to the study We could not find RCT or case–control studies that
design, the random effect models were used for combining definitively established the role of HD in PQ poisoning
the results. Calculating the τ2 = 0.00 in both meta‑analyses, outcomes (mortality/survival). Our results were according
the prediction interval was the same as the CIs. to data extracted from the cross‑sectional and case series

Figure 2: Random effects meta‑analysis (death events) (I2 = 0%, P value for heterogeneity was 0.386; Q = 1.146, τ2 = 0.00; P value for Egger’s test = 0.833)

Figure 3: Random effects meta‑analysis (death events) (I2 = 2%, P value for heterogeneity = 0.587; Q = 1.067, τ2 = 0.00, P value for Egger’s test = 0.698)

7 Journal of Research in Medical Sciences | 2022 |

 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

studies. Indeed, the qualities of clinical trial studies suggested that consuming more than 22.5 ml of 20% PQ
are not the same as observational studies. Therefore, can lead to a poor prognosis in patients with PQ poisoning.
their comparison may not be very accurate. However, This is in accordance with other studies.[29,54] Buckley[55] and
performing an RCT study with allocation concealment Afzali and Gholyaf[23] also showed that consuming around
has ethical concerns. Therefore, this review finding 10‑20 cc PQ could lead to fatal complications.
should be interpreted with caution until data from RCT or
case–control studies become available. Only in one study, the PQ intoxication severity and PQ
serum concentration based on nomogram and severity
Implications for practice index of PQ poisoning were considered in the selection of
Based on the findings, HD did not reduce the mortality in patients for performing HD.[9,56,57] In the other studies, organ
PQ poisoning cases. However, it should be mentioned that involvement might determine the severity of poisoning.
other variables, such as the amount of PQ ingested, vomiting The presence of esophageal and stomach ulceration within
after ingestion, time of ingestion to hospital admission, 24 h after ingestion of PQ shows severe toxicity.[29,58] In
performing early gastric lavage or administration of Delirrad et al.’s study,[29] five patients have undergone
activated charcoal, the severity of PQ intoxication based endoscopy, and all had high degrees of ulcerative lesions
on serum concentration, the interval of time to the start along the upper gastrointestinal tract. These patients died.
of HD, and duration of HD sessions, are crucial in the In Kavousi‑Gharbi’s study,[4] pulmonary fibrosis was
outcome (survival/mortality) and should be considered for observed in more than 55% of the patients who died. 14.8%
the decision to perform HD. of the patients who survived in the hospital had found
lung fibrosis. However, the data regarding the follow‑up
The time interval between PQ ingestion and the start of of the other patients have not been reported to determine
HD is an important issue. Previous studies demonstrated possible lung fibrosis and death. In Delirrad et al.’s study,[29]
that the initial hours after PQ ingestion was considered 16 patients from 19 dead patients had respiratory failure.
the optimal period to use extracorporeal elimination In Pavan’s study,[33] two patients survived. One patient
techniques such as HD.[17,50] Five patients in the study of ingested a half spoonful of PQ as suicide and the time of
Pavan[33] presented to the hospital within 1–5 days after ingestion to admission was 1 day. He received HD and
PQ ingestion and underwent HD because of AKI. HD was progressed to respiratory, renal, and hepatic involvement.
performed at least 6 h after admission in Kavousi‑Gharbi’s He survived, and his kidney function stabilized at serum
study.[4] Moreover, the time to HD was 6.73 h in Delirrad creatinine of 2 mg/dL. The other case accidentally ingested
et al.’s study.[29] This lag of time to HD may cause HD a few drops of PQ and was admitted to the hospital 5 h
efficacy to decrease. PQ can reach a plasma concentration after ingestion. He did not receive HD and survived. He
peak in 1 h due to rapid absorption[51] and accumulate in also developed respiratory and hepatic involvement.
targeted tissues. His kidney functions gradually improved to normal. He
received a hypoxic breathing mixture as a therapeutic
Moreover, there is a chance of re‑distribution from tissues modality. Furthermore, in this study, the follow‑up of the
to plasma. PQ distribution half‑life is around 5 h in humans, survived patients for evaluating lung fibrosis or death has
and about 6 h after its consumption, it reaches the maximum not been reported.
tissue concentration in the lungs.[52] Clinical features of PQ
poisoning and many organs’ cellular damage are primarily Limitations of the study
due to intracellular effects of PQ, such as generating reactive 1. The results regarding the follow‑up of the survived
oxygen species, lipid peroxidation, activation of NF‑kB, patients for pulmonary fibrosis and death had not been
mitochondrial damage, and apoptosis.[53] Therefore, late reported in the studies included in our meta‑analysis. It
admission to the hospital and delay in performing HD has been shown that pulmonary fibrosis occurs within
might have the reason for a worse outcome. 14 days[59]
2. We could not analyze data based on the time from
The amount of PQ ingested is another critical factor. In admission to start HD and the number of HD sessions
Delirrad et al.’s study,[29] the amount of PQ ingested in patients as the data were unavailable in some evaluated studies.
treated with HD was higher than those not receiving HD. Patients who presented later than 4 h may not benefit
Furthermore, the dose of PQ ingested in the nonsurvivors from HD.
was higher significantly from the survivor (290 ± 266 vs. 3. Although the main reason for early HD is the excretion
23.4 ± 22.3). Furthermore, Kavousi‑Gharbi et al.[4] reported of PQ from the body, this objective was not obvious in
that the amount of ingested PQ was higher in the patients all studies
who died than those who survived. 91.1% of the patients 4. The problem is an objective method of assessing risk
who had consumed more than 20 ml of PQ died. They in PQ poisoning is the measurement of the plasma

| 2022 | Journal of Research in Medical Sciences 8

 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

concentration concerning the time from ingestion. REFERENCES

However, plasma PQ assays are not available in
all centers. In addition, once a patient has been 1. Sabzghabaee AM, Eizadi‑Mood N, Montazeri K, Yaraghi A,
selected for HD, there is a further delay before the Golabi M. Fatality in paraquat poisoning. Singapore Med J
procedure commences because some laboratory tests 2010;51:496‑500.
2. Gunnell D, Eddleston M, Phillips MR, Konradsen F. The global
are needed for HD. Moreover, we now know that in
distribution of fatal pesticide self‑poisoning: Systematic review.
patients who present early after ingestion, plasma PQ BMC Public Health 2007;7:357.
concentrations decline extremely rapidly because of 3. Sandhu J, Dhiman A, Mahajan R, Sandhu P. Outcome of paraquat
distribution to tissues. It has been reported that HD is poisoning – A five year study. Indian J Nephrol 2003;13:64‑8.
used only as a supportive treatment for patients with 4. Kavousi‑Gharbi S, Jalli R, Rasekhi‑Kazerouni A, Habibagahi Z,
Marashi SM. Discernment scheme for paraquat poisoning:
AKI
A five‑year experience in Shiraz, Iran. World J Exp Med 2017;7:31‑9.
5. All important variables of outcome prediction such as 5. Dinis‑Oliveira RJ, Duarte JA, Sánchez‑Navarro A, Remião F,
the amount of ingested PQ, time from admission to start Bastos ML, Carvalho F. Paraquat poisonings: Mechanisms of
HD, and plasma PQ concentration had not been reported lung toxicity, clinical features, and treatment. Crit Rev Toxicol
in all studies included in the meta‑analysis. Therefore, 2008;38:13‑71.
we could not perform a subgroup meta‑analysis 6. Gawarammana IB, Buckley NA. Medical management of paraquat
ingestion. Br J Clin Pharmacol 2011;72:745‑57.
according to these variables.
7. Cherukuri H, Pramoda K, Rohini D, Thunga G, Vijaynarayana K,
Sreedharan N, et al. Demographics, clinical characteristics and
CONCLUSION management of herbicide poisoning in tertiary care hospital.
Toxicol Int 2014;21:209‑13.
In conclusion, according to our meta‑analysis, sufficient 8. Okonek S, Hofmann A, Henningsen B. Efficacy of gut lavage,
evidence did not exist from all included studies indicating hemodialysis, and hemoperfusion in the therapy of paraquat or
diquat intoxication. Arch Toxicol 1976;36:43‑51.
HD’s efficacy on the survival of patients with PQ poisoning.
9. Proudfoot AT, Prescott LF, Jarvie DR. Haemodialysis for paraquat
Given the importance of this intoxication, especially in a poisoning. Hum Toxicol 1987;6:69‑74.
developing country, meticulous and robust preventive 10. Cooke NJ, Flenley DC, Matthew H. Paraquat poisoning. Serial
strategies are suggested to be applied. Although HD did studies of lung function. Q J Med 1973;42:683‑92.
not affect the survival of patients, other variables such as 11. Mydlík M, Derzsiová K, Frank K. Renal replacement therapy in
the amount of ingested PQ, poisoning severity, the time acute poisonings‑one center experience. Przegl Lek 2013;70:381‑5.
12. Kang MS, Gil HW, Yang JO, Lee EY, Hong SY. Comparison between
between PQ ingestion and the start of HD, and duration
kidney and hemoperfusion for paraquat elimination. J Korean Med
and times of HD sessions may affect the outcome of the Sci 2009;24 Suppl: S156‑60.
patients with PQ poisoning. 13. Wang Y, Chen Y, Mao L, Zhao G, Hong G, Li M, et al. Effects
of hemoperfusion and continuous renal replacement therapy
Implications for research on patient survival following paraquat poisoning. PLoS One
More research is needed to determine the effectiveness of 2017;12:e0181207.
14. Koo JR, Kim JC, Yoon JW, Kim GH, Jeon RW, Kim HJ, et al. Failure
HD in patients with PQ poisoning.
of continuous venovenous hemofiltration to prevent death in
paraquat poisoning. Am J Kidney Dis 2002;39:55‑9.
Implications for clinician 15. Van de Vyver FL, Giuliano RA, Paulus GJ, Verpooten GA,
It may be practical to educate public health professionals Franke JP, De Zeeuw RA, et al. Hemoperfusion‑hemodialysis
and the general population about the fatal consequences ineffective for paraquat removal in life‑threatening poisoning? J
Toxicol Clin Toxicol 1985;23:117‑31.
of exposure to this toxic agent. It is also suggested to
16. Hampson EC, Pond SM. Failure of haemoperfusion and
replace the conventional HD with other extracorporeal haemodialysis to prevent death in paraquat poisoning.
interventions that may have beneficial effects. A retrospective review of 42 patients. Med Toxicol Adverse Drug
Exp 1988;3:64‑71.
Acknowledgments 17. Park S, Lee S, Park S, Gil H, Lee E, Yang J, et al. Concurrent
We would like to thank Prof. A. T. Proudfoot, Prof. M. hemoperfusion and hemodialysis in patients with acute pesticide
intoxication. Blood Purif 2016;42:329‑36.
Delirrad, and Dr. S. M. Marashi for their responses to our
18. Lee CJ, Hsu HW, Chang YL. Performance characteristics of
questions concerning their studies.[4,9,29] combined haemodialysis/haemoperfusion system for removal of
blood toxins. Med Eng Phys 1997;19:658‑67.
Financial support and sponsorship 19. Li C, Hu D, Xue W, Li X, Wang Z, Ai Z, et al. Treatment outcome
This study was supported by the Ethical Committee of of combined continuous venovenous hemofiltration and
Isfahan University of Medical Sciences (IR.MUI.MED. hemoperfusion in acute paraquat poisoning: A prospective
controlled trial. Crit Care Med 2018;46:100‑7.
REC.1399.512).
20. Hoffman S, Jedeikin R, Korzets Z, Shapiro AL, Kaplan R,
Bernheim J. Successful management of severe paraquat poisoning.
Conflicts of interest Chest 1983;84:107‑9.
There are no conflicts of interest. 21. Suntres ZE. Role of antioxidants in paraquat toxicity. Toxicology

9 Journal of Research in Medical Sciences | 2022 |

 Eizadi‑Mood, et al.: Hemodialysis efficacy on paraquat poisoning outcome

2002;180:65‑77. with cyclophosphamide for oral paraquat poisoning. Cochrane

22. Lin JL, Leu ML, Liu YC, Chen GH. A prospective clinical trial Database Syst Rev 2021;6:CD008084.
of pulse therapy with glucocorticoid and cyclophosphamide in 41. He F, Xu P, Zhang J, Zhang Q, Gu S, Liu Y, et al. Efficacy and
moderate to severe paraquat‑poisoned patients. Am J Respir Crit safety of pulse immunosuppressive therapy with glucocorticoid
Care Med 1999;159:357‑60. and cyclophosphamide in patients with paraquat poisoning:
23. Afzali S, Gholyaf M. The effectiveness of combined treatment A meta‑analysis. Int Immunopharmacol 2015;27:1‑7.
with methylprednisolone and cyclophosphamide in oral paraquat 42. Lin G, Long J, Luo Y, Wang Y, Zewu Q. Continuous
poisoning. Arch Iran Med 2008;11:387‑91. venovenous hemofiltration in the management of paraquat
24. Drault JN, Baelen E, Mehdaoui H, Delord JM, Flament F. Massive poisoning: A meta‑analysis of randomized controlled trials.
paraquat poisoning. Favorable course after treatment with Medicine (Baltimore) 2017;96:e6875.
n‑acetylcysteine and early hemodialysis. Ann Fr Anesth Reanim 43. Lan C, Lyu Q, Pei H, Meng X, Liu Q, Jia X, et al. Effect of
1999;18:534‑7. hemoperfusion combined with continuous veno‑venous
25. Hong SY, Hwang KY, Lee EY, Eun SW, Cho SR, Han CS, et al. Effect hemofiltration on acute paraquat poisoning: A Meta‑analysis.
of vitamin C on plasma total antioxidant status in patients with Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2018;30:783‑9.
paraquat intoxication. Toxicol Lett 2002;126:51‑9. 44. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group.
26. Dorooshi G, Zolfaghari S, Eizadi‑Mood N, Gheshlaghi F. A new Preferred reporting items for systematic reviews and meta‑analyses:
treatment approach for acute paraquat poisoning. J Res Pharm The PRISMA statement. PLoS Med 2009;6:e1000097.
Pract 2018;7:115‑6. 45. Jordan Z, Lockwood C, Munn Z, Aromataris E. The updated
27. Eizadi‑Mood N, Sabzghabaee AM, Yaraghi A, Montazeri K, Joanna Briggs institute model of evidence‑based healthcare. Int J
Golabi M, Sharifian A, et al. Effect of antioxidants on the outcome Evid Based Healthc 2019;17:58‑71.
of therapy in paraquat‑intoxicated patients. Trop J Pharm Res 46. Ma LL, Wang YY, Yang ZH, Huang D, Weng H, Zeng XT.
2011;10:27-31. Methodological quality (risk of bias) assessment tools for primary
28. Hedaiaty M, Sabzghabaee AM, Gheshlaghi F, Mood NE. Paraquat and secondary medical studies: What are they and which is better?
poisoning management in Iran (isfahan): Devising a protocol. Mil Med Res 2020;7:7.
J Adv Med Med Res 2016;10:1‑10.
47. Moola S, Munn Z, Tufanaru C, Aromataris E, Sears K, Sfetcu R,
29. Delirrad M, Majidi M, Boushehri B. Clinical features and prognosis
et al. Systematic reviews of etiology and risk. In: Joanna Briggs
of paraquat poisoning: A review of 41 cases. Int J Clin Exp Med
Institute Reviewer’s Manual. Ch. 7. The Joanna Briggs Institute;
2015;8:8122‑8.
2017. p. 5.
30. Weng CH, Chen HH, Hu CC, Huang WH, Hsu CW, Fu JF, et al.
48. DerSimonian R, Laird N. Meta‑analysis in clinical trials. Control
Predictors of acute kidney injury after paraquat intoxication.
Clin Trials 1986;7:177‑88.
Oncotarget 2017;8:51345‑54.
49. Egger M, Davey Smith G, Schneider M, Minder C. Bias in
31. Mood NE, Sabzghabaee AM, Ghodousi A, Yaraghi A, Mousavi A,
meta‑analysis detected by a simple, graphical test. BMJ
Massoum G, et al. Histo‑pathological findings and their
1997;315:629‑34.
relationship with age, gender and toxin amounts in paraquat
50. Gil HW, Hong JR, Jang SH, Hong SY. Diagnostic and therapeutic
intoxication. Pak J Med Sci 2013;29:403-408.2013.
approach for acute paraquat intoxication. J Korean Med Sci
32. Lheureux P, Leduc D, Vanbinst R, Askenasi R. Survival in a case
2014;29:1441‑9.
of massive paraquat ingestion. Chest 1995;107:285‑9.
51. Conning DM, Fletcher K, Swan AA. Paraquat and related
33. Pavan M. Acute kidney injury following Paraquat poisoning in
bipyridyls. Br Med Bull 1969;25:245‑9.
India. Iran J Kidney Dis 2013;7:64‑6.
52. Bismuth C, Scherrmann JM, Garnier R, Baud FJ, Pontal PG.
34. Ponce P, Lobos AV, Bordalo J, Moreira J. Treatment of paraquat
poisoning. Plasmapheresis versus hemodialysis. Acta Med Port Elimination of paraquat. Hum Toxicol 1987;6:63‑7.
1986;7:193‑6. 53. Gawarammana I, Buckley NA, Mohamed F, Naser K, Jeganathan K,
35. Li H, Song L, Jian X. Efficacy of hemoperfusion therapy in patients Ariyananada PL, et al. High‑dose immunosuppression to prevent
with paraquat poisoning: A meta‑analysis. Int J Clin Exp Med death after paraquat self‑poisoning – A randomised controlled
2017;10:11371‑81. trial. Clin Toxicol (Phila) 2018;56:633‑9.
36. Nasr Isfahani S, Farajzadegan Z, Sabzghabaee AM, Rahimi A, 54. Senarathna L, Eddleston M, Wilks MF, Woollen BH, Tomenson JA,
Samasamshariat S, Eizadi‑Mood N. Does hemoperfusion in Roberts DM, et al. Prediction of outcome after paraquat poisoning
combination with other treatments reduce the mortality of by measurement of the plasma paraquat concentration. QJM
patients with paraquat poisoning more than hemoperfusion alone: 2009;102:251‑9.
A systematic review with meta‑analysis. J Res Med Sci 2019;24:2. 55. Buckley NA. Pulse corticosteroids and cyclophosphamide in
37. Xu YG, Lu YQ. Systematic review and meta‑analysis of the paraquat poisoning. Am J Respir Crit Care Med 2001;163:585.
efficacy and safety of immunosuppressive pulse therapy in 56. Bismuth C, Garnier R, Baud FJ, Muszynski J, Keyes C. Paraquat
the treatment of paraquat poisoning. J Zhejiang Univ Sci B poisoning. An overview of the current status. Drug Saf
2019;20:588‑97. 1990;5:243‑51.
38. Eddleston M, Wilks MF, Buckley NA. Prospects for treatment of 57. Wunnapuk K, Mohammed F, Gawarammana I, Liu X, Verbeeck RK,
paraquat‑induced lung fibrosis with immunosuppressive drugs Buckley NA, et al. Prediction of paraquat exposure and toxicity in
and the need for better prediction of outcome: A systematic review. clinically ill poisoned patients: A model based approach. Br J Clin
QJM 2003;96:809‑24. Pharmacol 2014;78:855‑66.
39. Agarwal R, Srinivas R, Aggarwal A, Gupta D. Immunosuppressive 58. Bismuth C, Garnier R, Dally S, Fournier PE, Scherrmann JM.
Therapy in Lung Injury Due to Paraquat Poisoning: Prognosis and treatment of paraquat poisoning: A review of
A Meta‑Analysis. Database of Abstracts of Reviews of 28 cases. J Toxicol Clin Toxicol 1982;19:461‑74.
Effects (DARE): Quality‑Assessed Reviews: Centre for Reviews 59. Li S, Li H, Zheng Z, Dong H. Pathologic characteristics of diffuse
and Dissemination (UK); 2007. alveolar damage induced by paraquat in rat lungs. Hua Xi Yi Ke
40. Li LR, Chaudhary B, You C, Dennis JA, Wakeford H. Glucocorticoid Da Xue Xue Bao 1994;25:337‑40.

| 2022 | Journal of Research in Medical Sciences 10

Supplementary Table 1: Specify the exact keywords searched in each database
Database Title search Subject search
WOS TITLE: ((haemodialysis OR dialysis or “Renal Dialysis” OR TOPIC: ((haemodialysis OR dialysis or “Renal Dialysis”
hemodialysis) AND (paraquat OR gramoxone) AND (poisoning OR hemodialysis) AND (paraquat OR gramoxone) AND
OR intoxication OR overdose OR toxicity) AND (survival OR (poisoning OR intoxication OR overdose OR toxicity)
death OR mortality OR fatality)) AND (survival OR death OR mortality OR fatality))
Scopus TITLE ((haemodialysis OR dialysis OR “Renal Dialysis” OR KEY ((haemodialysis OR dialysis OR “Renal Dialysis” OR
hemodialysis) AND (paraquat OR gramoxone) AND (poisoning hemodialysis) AND (paraquat OR gramoxone) AND (poisoning
OR intoxication OR overdose OR toxicity) AND (survival OR OR intoxication OR overdose OR toxicity) AND (survival OR
death OR mortality OR fatality)) death OR mortality OR fatality))
Proquest Ti ((haemodialysis OR dialysis OR “Renal Dialysis” OR Su ((haemodialysis OR dialysis OR “Renal Dialysis” OR
hemodialysis) AND (paraquat OR gramoxone) AND (poisoning hemodialysis) AND (paraquat OR gramoxone) AND (poisoning
OR intoxication OR overdose OR toxicity) AND (survival OR OR intoxication OR overdose OR toxicity) AND (survival OR
death OR mortality OR fatality))) death OR mortality OR fatality))
Science Title: ((haemodialysis OR dialysis OR “Renal Dialysis” OR Title, abstract, keyords: ((haemodialysis OR dialysis OR
direct hemodialysis) AND (paraquat OR gramoxone) AND (poisoning “Renal Dialysis” OR hemodialysis) AND (paraquat OR
OR intoxication OR overdose OR toxicity) AND (survival OR gramoxone) AND (poisoning OR intoxication OR overdose OR
death OR mortality OR fatality)) toxicity) AND (survival OR death OR mortality OR fatality))
Springer Title: ((haemodialysis OR dialysis OR “Renal Dialysis” OR With all of the words: ((haemodialysis OR dialysis OR “Renal
hemodialysis) AND (paraquat OR gramoxone) AND (poisoning Dialysis” OR hemodialysis) AND (paraquat OR gramoxone)
OR intoxication OR overdose OR toxicity) AND (survival OR AND (poisoning OR intoxication OR overdose OR toxicity)
death OR mortality OR fatality)) AND (survival OR death OR mortality OR fatality))
PubMed (haemodialysis[Title] OR dialysis[Title] OR “Renal (haemodialysis OR dialysis or “Renal Dialysis” OR
Dialysis”[Title] OR hemodialysis[Title]) AND (paraquat[Title] OR hemodialysis) AND (paraquat OR gramoxone) AND (poisoning
gramoxone[Title]) AND (poisoning[Title] OR intoxication[Title] OR intoxication OR overdose OR toxicity) AND (survival OR
OR overdose[Title] OR toxicity[Title]) AND (survival[Title] OR death OR mortality OR fatality) [MeSH terms]
death[Title] OR mortality[Title] OR fatality[Title])
Embase Title: ((haemodialysis OR dialysis OR “Renal Dialysis” OR EMTree: ((haemodialysis OR dialysis OR “Renal Dialysis”
hemodialysis) AND (paraquat OR gramoxone) AND (poisoning OR hemodialysis) AND (paraquat OR gramoxone) AND
OR intoxication OR overdose OR toxicity) AND (survival OR (poisoning OR intoxication OR overdose OR toxicity)
death OR mortality OR fatality)) AND (survival OR death OR mortality OR fatality))
Cochrane (haemodialysis OR dialysis OR “Renal Dialysis” OR (haemodialysis OR dialysis OR “Renal Dialysis” OR
library hemodialysis) AND (paraquat OR gramoxone) AND (poisoning hemodialysis) AND (paraquat OR gramoxone) AND (poisoning
OR intoxication OR overdose OR toxicity) AND (survival OR OR intoxication OR overdose OR toxicity) AND (survival OR
death OR mortality OR fatality) in Record Title death OR mortality OR fatality) in keyword

Supplementary Table 2: Quality assessment of

cross‑sectional studies based on the Joanna Briggs
Institute critical appraisal checklist
Studies Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Score
Proudfoot A.T.[9] Yes Yes Yes Yes No No Yes Yes 6/8
Delirrad M.[29] Yes Yes No Yes No No Yes Yes 5/8
Kavousi‑Gharbi S.[4] Yes Yes Yes Yes No No Yes Yes 6/8
Q1. Were the criteria for inclusion in the sample clearly defined?; Q2. Were the study
subjects and the setting described in detail?; Q3. Was the exposure measured in a
valid and reliable way?; Q4. Were objective, standard criteria used for measurement
of the condition?; Q5. Were confounding factors identified?; Q6. Were strategies to
deal with confounding factors stated?; Q7. Were the outcomes measured in a valid
and reliable way?; Q8. Was appropriate statistical analysis used?
Supplementary Table 3: Quality assessment of randomized trial studies based on the Joanna Briggs Institute critical
appraisal checklist
Studies Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Score
Ponce P.[34] Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes Yes 10/13
Q1. Was true randomization used for assignment of participants to treatment groups?; Q2. Was allocation to treatment groups concealed?; Q3. Were treatment groups similar
at the baseline?; Q4. Were participants blind to treatment assignment?; Q5. Were those delivering treatment blind to treatment assignment?; Q6. Were outcomes assessors
blind to treatment assignment?; Q7. Were treatment groups treated identically other than the intervention of interest?; Q8. Was follow up complete and if not, were differences
between groups in terms of their follow up adequately described and analyzed?; Q9. Were participants analyzed in the groups to which they were randomized?; Q10. Were
outcomes measured in the same way for treatment groups?; Q11. Were outcomes measured in a reliable way?; Q12. Was appropriate statistical analysis used?; Q13. Was the
trial design appropriate, and any deviations from the standard RCT design (individual randomization, parallel groups) accounted for in the conduct and analysis of the trial?.
RCT=Randomized clinical trial

Supplementary Table 4: Quality assessment of case

series studies based on the Joanna Briggs Institute
critical appraisal checklist
Studies Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Score
Paven M.[33] Yes Yes Yes No Yes No Yes Yes No Yes 7/10
Cook NJ.[10] Yes Yes Yes No Yes No Yes Yes No Yes 7/10
Q1. Were there clear criteria for inclusion in the case series?; Q2. Was the condition
measured in a standard, reliable way for all participants included in the case series?;
Q3. Were valid methods used for identification of the condition for all participants
included in the case series?; Q4. Did the case series have consecutive inclusion
of participants?; Q5. Did the case series have complete inclusion of participants?;
Q6. Was there clear reporting of the demographics of the participants in the study?;
Q7. Was there clear reporting of clinical information of the participants?; Q8. Were
the outcomes or follow up results of cases clearly reported?; Q9. Was there clear
reporting of the presenting site (s)/clinic (s) demographic information?; Q10. Was
statistical analysis appropriate?