Risk factors for Encapsulating Peritoneal Sclerosis in patients undergoing peritoneal dialysis: A meta-analysis.

PURPOSE: Encapsulating Peritoneal Sclerosis (EPS) is the most serious complication of long-term peritoneal dialysis (PD), which considerably reduces the patient's quality of life, leading to patients discontinuing PD. Considering these negative effects, it is necessary to systematically review and determine the risk factors of EPS.
METHODS: The PubMed, Embase, Web of Science, Cochrane Library, and China Biology Medicine (CBM) were searched from their inception to January 1st, 2022, and the bibliographies from the citations of relevant articles were manually searched. The ROBINS-I (Risk of Bias in Non-randomized studies of Interventions) tool was used to evaluate the risk of bias of included studies.
RESULTS: Ten studies involving 12595 participants were included in this meta-analysis. The results revealed that a younger age at PD onset (MD = -7.70, 95% CI, -11.53~-3.86), a higher transporter (MD = 0.13, 95% CI, 0.09~0.18), a longer PD duration (SMD = 1.15, 95% CI, 0.68~1.61), a longer peritonitis duration (MD = 12.66, 95% CI, 3.85~21.47), and history of glomerulonephritis (OR = 1.42, 95% CI, 1.02~1.97) were significant risk factors for EPS. However, sex, use of icodextrin, the number of peritonitis episodes, and history of multicystic kidney disease did not affect the risk of EPS.
CONCLUSIONS: This review provides a scientific basis for further understanding the etiology of PD-related EPS and improving prevention strategies. More high-quality studies are necessary to validate this paper's findings.

Introduction

Encapsulating Peritoneal Sclerosis (EPS), also known as Peritoneal Fibrosis and Sclerosing Encapsulating Peritonitis, is a rare but the most severe complication associated with peritoneal dialysis, especially in patients who are on long-term PD [1]. In 2000, the International Society for Peritoneal Dialysis has described EPS as a rare clinical syndrome characterized by intermittent, persistent or recurrent intestinal obstruction caused by diffuse adhesions of a thickened and sclerosing peritoneum [2]. If obstruction does not resolve, intestinal mural ischemia and peritoneal adhesions may further develop, resulting in infections, bloody dialysate, systemic inflammatory response syndrome, sepsis, and death [3]. Adverse outcomes of PD-related EPS are diverse, including loss of incipient peritoneal ultrafiltration capacity, a higher transport status, and a lower residual kidney function [4]. Although the underlying mechanism by which normal peritoneal membrane progresses to peritoneal fibrosis in PD patients is not yet clear, overexpression of a cascade of proinflammatory cytokines (e.g. transforming growth factor β1, interleukin-6) and profibrotic gene (e.g. Fibronectin 1, ɑ smooth muscle actin), neoangiogenesis, and mesothelial to mesenchymal transition (MMT) may be involved in the occurrence and progression of EPS [5-7]. Unfortunately, the optimal treatment for EPS is not clear. There are no standard therapies or randomized controlled trials to cure EPS patients but some empiric knowledge based mainly on observational evidences include surgical and medical interventions, might be beneficial [8]. Generally, PD should be discontinued after the diagnosis of EPS is made. Additional therapeutic strategies include steroid therapy and other immunosuppression, nutritional support and assessment, tamoxifen, parenteral or enteral nutrition, and peritoneal lavage [8, 9]. Despite these, the mortality rates of patients with EPS are still 25%-55% in adults [10-12], with positively correlated with the duration of PD therapy.

The overall incidence of EPS obtained from a variety of countries is low and varies between 0.5% and 7.3% [13-18]. A registry study from Australian reported that the incidence rate increased with duration of PD therapy [14]. Similar results were obtained in a prospective study from Japan, in which the incidence of EPS was 0.7%, 2.1%, 5.9%, 17.2% in patients who had been on PD for more than 5, 8, 10, and 15 years, respectively [13]. Many factors including duration of PD, age at initiation of PD, bio-incompatible dialysate, high-transporter membrane, duration of UFF (ultrafiltration failure), duration of peritonitis, number of peritonitis episodes, and kidney transplantation may lead to EPS [1, 17, 19, 20]. However, both a single-center retrospective study from Taiwan and dual-center retrospective study from Iran analysed the relationship between age at PD onset and the development of EPS, with the latter concluding that younger age was a high-risk factor for EPS [21], while the former found no significant difference according to age [22]. The question, therefore, ‘What are the risk factors related to EPS in PD patients?’ remains unanswered.

Regarding controversial results of existing studies, a meta-analysis was conducted to identify comprehensive results which could contribute to develop clinical strategies for early prevention of EPS. To our knowledge, this is the first meta-analysis to analysis the risk factors associated with EPS.

Methods

This meta-analysis was reported in line with PRISMA guidelines [23] and registered with the International Prospective Register of Systematic Reviews (PROSPERO) [24] with registration number (CRD42022302786).

Search strategy

On 10 January 2022, two independent reviewers (Yuanyuan Li and Hanxu Zeng) searched relevant citations in PubMed, Embase, Web of Science, Cochrane Library, and China Biology Medicine (CBM) from their inception to January 1st, 2022, and the bibliographies from the citations of relevant articles were manually searched. We also searched the http://www.greylit.org/ website for studies that were registered as completed but not yet published. There were no restrictions on language with limited to human subjects. We used a combination of search medical subject heading (MeSH) related to EPS and risk factor. Any discrepancies regarding study selection were resolved by additional assessment by a third investigator (Yonggui Wu). S2 File outlined the detailed search strategy of PubMed.

Inclusion and exclusion criteria

The criteria for original research to be included in this meta-analysis were as follows: (1) This review included case-control studies, cross-sectional studies cohort studies, and randomized controlled trials (RCTs) to explore the risk factors associated with EPS, (2) The studies were conducted in patients whose peritoneal dialysis lasted three months at least and age >18years, (3) Availability of complete data required for the analysis, and (4) Clear and consistent diagnosis of EPS based on the diagnostic criteria. The exclusion criteria were (1) Reviews, conference papers, lectures, drug experiments, animal studies; (2) Studies that included duplicated data; (3) Primary data could not been extracted.

Data extraction and risk assessment

Two authors (Yuanyuan Li and Hanxu Zeng) searched and extracted the data of the included primary studies, aggregating the resulting data into a structured table: first author’s name, study duration, study type, country/region, proportion of women, sample size, and risk of bias. The ROBINS-I (Risk of Bias in Non-randomized studies of Interventions) tool was used by two authors (Yuanyuan Li and Hanxu Zeng) to evaluate the risk of bias of included studies based on the following seven domains: bias due to confounding; bias due to selection of participants; bias in classification of interventions; bias due to deviations from intended interventions; bias due to missing data; bias in measurement of outcomes; bias in selection of the reported result. This tool categorizes the risk of bias as low, moderate, serious, critical, and unclear [25]. Disagreements were resolved by consultation with a third reviewer (Yonggui Wu).

Statistical analysis

All statistics were collated and analyzed using the Review Manager 5.3 and Stata 14.0 software. Continuous variables were summarized using mean difference (MD) or standardized mean difference (SMD) with their 95% confidence intervals (CI), For dichotomous variables, odds ratio (OR) with their 95% CI were estimated. In addition, heterogeneity was quantified using the Q test and I2 statistics. When I2 was < 50% and Q chisquared test result > 0.1, it shows that there is no large heterogeneity among the trials and the fixed-effect model was used; else a random-effect model was used. When large heterogeneity was present, sensitivity analysis were performed to identify responsible outlier studies. Furthermore, publication bias was further evaluated using the Egger’s test. A trial sequence analysis (TSA) was used to analyze the sample size required for this meta-analysis to improve the credibility of this study.

SPSS Statistics version 25.0 (SPSS INC., Chicago, IL) was used to analyze the data of kappa of agreement during literature search. The data of kappa >0.60 were considered to indicate good agreement.

Results

Search results and study characteristics

A total of 495 potentially relevant articles were retrieved initially, in which 371 remained after duplicates from different databases were deleted. Next, 339 articles not aligned with the inclusion criteria were excluded after reading the titles and abstracts. Subsequently, 32 papers meeting the inclusion criteria were considered, in which 22 were excluded due to duplicate records, unadjusted confounding factors and inconsistent research design. Ultimately, we included 10 standards-compliant articles (Fig 1). The data of kappa of agreement during the systematic searches were 0.721, indicating good agreement.

Fig 1

Flow diagram of the literature selection in this meta-analysis.

Eight retrospective case-control studies and two prospective cohort studies [13, 15, 21, 22, 26–31] included in this meta-analysis were published between 2004 and 2018, with a total of 12595 patients on PD, which included 293 patients with EPS and 12302 with no EPS. Excluding Yamamoto et al.’s study [30], the number of participants in other studies ranged from 384 to 7618, and the incidence of EPS ranged from 0.4% to 11.7%. Study duration varied from 4 years to 35 years. In addition, two out of 10 studies had a women predominant (range of 52–53%). These studies were conducted in Iran, Japan, Netherlands, Taiwan, Australia, New Zealand, Turkey, and Ireland. The diagnosis of EPS in these ten articles was established based on the guidelines developed by the ISPD including clinical diagnosis, radiologic diagnosis and pathologic diagnosis. The basic characteristics of the selected articles are presented in Table 1. Potential bias was evaluated by the ROBINS-I tool for observational studies. All studies presented a moderate overall risk of bias, in which confounding bias remains in all studies due to their non-randomized nature. The exact details of the risk evaluation are summarized in S1 Table.

Table 1

Basic characteristics of all studies in this meta-analysis.

Author	Study duration	Study design	Country/Region	Women(%)	Age range		Sample sizes		Risk of biasa
					EPS	NO EPS
Alatab, S. [21]	1995–2012	Case control	Iran	53	32.75±10.8	49.61±16.2	case:12	control: 122	Moderate
Nakao, M. [29]	1989–2010	Case control	Japan	33	54.6±11.5	56.8±13.0	case:50	control:57	Moderate
Nakao, M. [28]	1980–2015	Case-control	Japan	27	44±12	55±15	case:44	control:659	Moderate
Yamamoto, R. [30]	NR	Case control	Japan	33	48.9±12.8	52.6±10.4	case:18	control:60	Moderate
Korte, M. R. [27]	1996–2007	Prospective cohort	Netherlands	34	34.7±15.4	51.5±14.7	case:63	control:126	Moderate
Hsu, H. J. [22]	1990–2014	Case control	Taiwan	44	54.3±18.27	53.7±17.6	case:6	control:559	Moderate
Johnson, D. W. [15]	1995–2007	Case control	Australia and New Zealand	49	49.1±12.8	58.0±16.6	case:33	control:7585	Moderate
Koc Y. [26]	2001–2016	Case control	Turkey	52	40.8±14.6	45.9±16.6	case:26	control:358	Moderate
Kawanishi, H. [13]	1999–2003	Prospective cohort	Japan	38	54.7±8.9	56.7±14.1	case:48	control:453	Moderate
Phelan, PJ. [31]	1989–2008	Case control	Ireland	42	43.3±14.4	50.2±14.1	case:11	control:604	Moderate

a Risk of bias was evaluated using ROBINS-I tool.

Synthesis of the results

More than three studies involving identically defined risk factors for EPS were pooled in the meta-analysis. A total of 9 potential risk factors including age at PD onset, sex, duration of PD, dialysate-to-plasma creatinine ratio (D/P Cr), number of peritonitis episodes, duration of peritonitis, use of icodextrin, history of glomerulonephritis (GN), and history of polycystic kidney disease (PKD) were identified. The details are shown in Table 2. Heterogeneity was found in potential risk factors of exposure to age at PD onset, duration of PD, D/P Cr, peritonitis duration, and use of icodextrin. Therefore, the combined effect sizes of these factors were calculated using a random-effects model.

Table 2

Meta-analysis of the pooled risk factors for EPS.

Risk factors	Studies	Model	Heterogeneity test I2(%) P		OR/MD/SMD	95% CI	P-value
Age on dialysis	10	Random	82	<0.001	-7.70	(-11.53~ -3.86)	<0.001
Sex	9	Fixed	0	0.98	1.16	(0.89–1.52)	0.26
D/P Cr	3	Random	72	0.03	0.13	(0.09–0.18)	<0.001
PD duration	7	Random	87	<0.001	1.15	(0.68–1.61)	<0.001
Peritonitis duration	3	Random	54	0.11	12.66	(3.85–21.47)	0.005
Peritonitis history	3	Fixed	35	0.21	1.59	(0.66–3.84)	0.30
icodextrin	3	Random	96	<0.001	1.09	(0.08–15.40)	0.95
GN	6	Fixed	34	0.18	1.42	(1.02–1.97)	0.04
PKD	6	Fixed	0	0.61	0.68	(0.31–1.46)	0.32

Abbreviations: D/P Cr, dialysate-to-plasma creative ratio; GN, glomerulonephritis; PKD, polycystic kidney disease; OR, odds ratio; MD, mean difference; SMD, standardized mean difference.

Demographic factors for the occurrence of EPS

Age at PD onset

All studies that met the inclusion criteria reported on the relationship between age and EPS in a total of 10894 patients (311 patients with EPS and 10583 patients without EPS). Due to the obvious heterogeneity (P < 0.001, I2 = 82%) (Fig 2A), the random-effects model was applied, and the final results showed that significant statistical difference was found between age at PD onset and the development of EPS (MD = -7.70, 95% CI, -11.53~-3.86, P < 0.0001).

Fig 2

A. Forest plot of age at PD onset and EPS in the exploratory meta-analysis. The mean difference and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. B. Forest plot of sex and EPS in the exploratory meta-analysis. The odds ratio and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. C. Forest plot of D/P Cr and EPS in the exploratory meta-analysis. The mean difference and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right.

Sex

A fixed-effect model was utilized to analyze the data from 9 articles (P = 0.98, I2 = 0%), with a total of 10279 patients (5645 male and 4634 female patients). The comprehensive results suggested that in terms of sex, no statistical difference was found between the EPS patients and non-EPS patients (OR = 1.16, 95% CI, 0.89–1.52, P = 0.26) (Fig 2B).

D/P Cr

Three articles [28-30] provided reliable data to study the level of D/P Cr in a total of 888 patients (112 EPS and 776 non-EPS patients). Since the data analysis results showed considerable heterogeneity (P = 0.03, I2 = 72%), a random-effects model was used to analyze the combined MD with a consequence of 0.13 (95% CI, 0.09–0.18, P < 0.001) (Fig 2C).

Duration of PD

Seven articles [13, 21, 22, 26, 27, 29, 30] reported the relevant data, with a total of 1958 patients (223 EPS and 1735 non-EPS patients), among which we found significant heterogeneity (P < 0.001, I2 = 87%) (Fig 3A). Ultimately, the results of the random-effects model obtained showed that the risk of EPS was higher in patients who have experienced PD for a longer time (SMD = 1.15, 95% CI, 0.68–1.61, P < 0.001).

Fig 3

A. Forest plot of PD duration and EPS in the exploratory meta-analysis. The standard mean difference and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. B. Forest plot of number of peritonitis episodes and EPS in the exploratory meta-analysis. The odds ratio and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. C. Forest plot of peritonitis duration and EPS in the exploratory meta-analysis. The mean difference and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right.

Number of peritonitis episodes

Three studies [21, 29, 30] reported the impact of peritonitis on EPS, with a total of 319 patients (80 EPS and 239 non-EPS patients). Due to the low heterogeneity, a fixed-effects model was used (P = 0.21, I2 = 35%). Notwithstanding, the final results did not describe the correlation between the number of episodes of peritonitis and the occurrence of EPS (OR = 1.59, 95% CI, 0.66–3.84, P = 0.30) (Fig 3B).

Peritonitis duration

A random-effects model was applied to evaluate the impact of peritonitis duration on EPS (P = 0.11, I2 = 54%), with a total of 1375 patients (100 EPS and 1275 non-EPS patients). Three articles [22, 28, 29] were analyzed, and there was clear evidence that in patients with PD, the duration of peritonitis was a high-risk factor influencing the occurrence and development of EPS (MD = 12.66, 95% CI, 3.85–21.47, P = 0.005) (Fig 3C).

Use of icodextrin

Analyzing the data obtained in 3 articles [27-29], with a total of 999 patients (321 experimental and 678 control group), a random-effects model was used because of obvious heterogeneity (P < 0.001, I2 = 96%) (Fig 4A). The final results suggested no statistical difference in the combined effect size (OR = 1.09, 95% CI, 0.08–15.40, P = 0.95).

Fig 4

A. Forest plot of use of icodextrin and EPS in the exploratory meta-analysis. The odds ratio and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. B. Forest plot of GN and EPS in the exploratory meta-analysis. The odds ratio and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right. C. Forest plot of PKD and EPS in the exploratory meta-analysis. The odds ratio and 95% CI for each study and the final combined results are displayed numerically on the left and graphically as a forest plot on the right.

Glomerulonephritis (GN)

Six articles [15, 21, 26–29] reported a link between a history of glomerulonephritis and EPS, with a total of 9135 patients (2413 patients with glomerulonephritis and 6722 patients with no glomerulonephritis). A fixed-effects model was used for data analysis (P = 0.18, I2 = 34%). The results suggested that patients with glomerulonephritis were more likely to develop EPS (OR = 1.42, 95% CI, 1.02–1.97, P = 0.04) (Fig 4B).

Polycystic kidney disease (PKD)

After analyzing the data, 6 articles [15, 21, 26–29] were included in this analysis (P = 0.61, I2 = 0%), with a total of 9135 patients (420 PKD patients and 8715 non-PKD patients). Subsequent results obtained using the fixed-effects model suggested that there was no statistical difference between the history of PKD and EPS (OR = 0.68, 95% CI, 0.31–1.46, P = 0.32) (Fig 4C).

Other risk factors

By browsing these 8 selected studies, we found that serum creatine, serum calcium, serum phosphorus, serum albumin, hemoglobin, β2M (β2 microglobulin), systolic blood pressure, diastolic blood pressure, lupus erythematosus, APD (automated PD), HD (hemodialysis), fungus related peritonitis, CAD (coronary artery disease), etc, were mentioned in 2 studies. Furthermore, ducation, smoking history, history of dialysis, BMI (body mass index), weight, CVA (cerebrovascular accident), chronic lung disease, peripheral vascular disease, use of hypertonic solution, catheter site infection, immunoglobulin A nephropathy, renovascular disease, reflux nephropathy, hemolytic uremic syndrome, hypertension, collagen vascular disease, cancer, etc, were mentioned in one study. Considering the small number of studies involved and inadequate sample size, a large sample of clinical data is still necessary for further confirmation. Hence, the meta-analysis for these variables is not available here.

We performed a TSA of EPS-related risk factors, including age at PD onset, D/P Cr, peritonitis duration, and history of GN. Regarding the differences in data between different studies, the TSA on PD duration is not available here. As shown in S1 and S2 Figs, the Z-curves cross both the conventional boundaries as well as O’Brien-Fleming alpha-spending boundaries, indicating the sample size is sufficient to get significant results.

Sensitivity analysis

Sensitivity analysis was performed to analyze the impact of each study on the net results. In our study, the pooling effect was no significant difference before and after excluding each study with high heterogeneity one by one, indicating the results of sensitivity analysis are stable (S3 Fig).

Publication bias

As is shown in S2 Table, the results of the Egger’s test for the analysis of EPS-related risk factors among PD patients showed that the included studies may have no possibility of publication bias (P>0.05).

Discussion

EPS, known for its high mortality, is one of the most serious complications of PD. Although many studies have reported risk factors for the occurrence of EPS, the data remains controversial. To the best of our knowledge, this is the first meta-analysis summarizing the risk factors of EPS. To accurately determine the risk factors related to the occurrence of EPS, we collected the relevant data from ten observational studies in eight countries or regions. Differences are found in the incidence of EPS reported in various reports (from 0.4% to 11.7%). It is worth noting that 2 important factors may influence this result. First, the eight studies included in this study are retrospective study designs, which may raise the possibility of recall bias. Second, the use of PD fluid in the studies differed slightly. For example, conventional PD fluid, biocompatible PD fluid and icodextrin were used in the report of Nakao et al, instead, all the patients used standard Dianeal dialysis solutions in the report of Alatab et al. Finally, lower age at PD onset, higher D/P Cr, longer duration of PD, longer peritonitis duration, and presence of GN were identified as significant risk factors for EPS.

This study demonstrated that age was a risk factor for the development of EPS, wherein the age of patients with EPS was lower than that of the non-EPS group (MD = -7.70, 95% CI, -11.53~-3.86, P < 0.0001). Similarly, a retrospective study showed that the average age at PD onset of 12 patients with EPS (35.1 ± 3.3 years) was significantly lower than that of patients on PD without EPS (47.3 ± 1.1 years), and the difference was as much as 12 years [30]. Given its independent importance, this variable is not only a consequence of longer duration on PD or longer follow-up after PD onset. Mesothelial cells play a key role in the process of peritoneal remodelling [32]. Mesothelial to mesenchymal transition is a main mechanism of peritoneal fibrosis in which mesothelial cells secrete cytokines during exposure to multiple insults (e.g. glucose) that subsequently result in the recruitment of fibroblasts and macrophages [20, 33]. Perhaps this disrupted peritoneal remodeling function is more vigorous in younger patients, which might be the main reason for the earlier development of fibrosis.

D/P Cr is an international index for evaluating the state of peritoneal transport. In the studies reported by Yamamoto et al. [19, 30], they defined a D/P Cr greater than 0.75 as a higher peritoneal transport state, which was found to be an important independent risk factor for EFS. Therefore, we sought to elucidate whether an increased peritoneal D/P Cr ratio might serve as an indicator of the early development of EPS. This study revealed that the value of D/P Cr was higher in the EPS group than in the non-EPS patient group, which indicates that the peritoneal transport status affects the incidence of EPS [34-36]. The development of a high peritoneal transport state is multi-factorial, partly because of the changes in the morphology of the capillaries supplying the peritoneum, and the effects of some humoral factors, such as angiogenic factors, cannot also be ruled out [37, 38]. These changes would ultimately trigger peritoneal sclerosis. More research, therefore, is necessary to focus on the relationship between high transport state and EPS in the future.

The results of our meta-analysis showed that the duration of PD was positively correlated with the incidence of EPS (SMD = 1.15, 95% CI, 0.68–1.61, P < 0.001). PD duration has been regarded as the most important risk factor for EPS development in several studies [3, 39–41]. For example, in a retrospective study, the overall incidence of EPS was only 0.3% (1/353) within the first four years of PD, whereas this figure reached 4.2% (3/71), 41.6% (5/12) and 27.3% (3/11) for patients with PD therapy for 4–6, 6–8 and >8years, respectively [21]. Similarly increase in EPS incidence also reported by other studies [12–14, 16, 27, 40–42]. Long-term PD can affect the peritoneal function and increase the number of peritonitis episodes, which might partly account for this phenomenon. For high-risk patients, discontinuing PD after 8 years might be an effective measure to prevent the occurrence of EPS [8].

Although peritonitis episodes particularly caused by fungal or bacterial organisms have long been regarded as a significant risk factor for EPS development [14, 30, 43–45], contradictory studies also exist. A Japanese prospective study reported that 25% of patients who developed EPS were related bacterial peritonitis [13], whereas other studies could not showed an associated between peritonitis episodes and EPS [4, 12, 27, 46–48]. This meta-analysis also indicated that the number of peritonitis episodes might not be relevant to EPS development. The ISPD, however, suggested that patients on PD with severe and/or non-resolving peritonitis have a higher tendency to develop EPS [2]. Yamamote and colleagues explained this result, suggesting that the occurrence of peritonitis-induced EPS might be affected by the duration of PD, thereby exaggerating the risk assessment between the two [30]. This conclusion needs to be further confirmed in future studies. Interestingly, previous studies have emphasized that it is the duration of peritonitis that increases the risk of EPS [21, 22, 28–30, 49]. This association can provide valuable information for the early detection and treatment of EPS.

Our study also found that the risk of EPS occurrence was 1.42 times higher in GN patients than in PD patients without GN. As many articles have reported, Tacrolimus and cyclosporine can promote fibrosis progression possibly by activating the epithelial-mesenchymal transition(EMT) response (increased transforming growth factor β1 and ɑ smooth muscle actin expression) and increasing the expression of fibrose-related genes [50-52]. Nevertheless, there was no information about previous drugs used in the glomerulonephritis group. This should be one of the present study limitations. Hence, before drawing definitive conclusions, additional research is necessary.

The advantages of this study are adapting comprehensive search, including large sample size studies and assessing the specific risk factors using the meta-analytic method. Nevertheless, several limitations of this study must be considered when interpreting the results. First, we included a total of 10 papers in our meta-analysis, of which 4 papers originated from Japan; this could be a reason for bias. Second, medication history was not provided in these literatures, which might affect outcomes of patients. Finally, the validity of statistical analyses was affected by the disparities in heterogeneity among studies. Due to the limited number of studies, we only performed sensitivity analysis without subgroup analysis. Nevertheless, our results made an important contribution to the identification of well-recognized risk factors associated with EPS by integrating studies involving risk factors for EPS.

Conclusions

In conclusion, EPS is a rare and most serious complication in patients on long-term PD. This study comprehensively summarized EPS-related risk factors and concluded that lower age at PD onset, higher D/P Cr, longer duration of PD, longer peritonitis duration, and history of GN are associated with significantly increased risk of EPS among PD patients. These findings provide a scientific basis for further understanding the etiology of PD-related EPS and improving prevention strategies. Particularly, the incidence of EPS can be reduced by controlling a series of modifiable risk factors. More high-quality studies are needed to validate this paper’s findings in the future.

TSA results of the related risk factors.

(A) Age at PD onset; (B) D/P Cr. The information size was calculated based on MD, alpha of 5%, power of 80%.

(TIF)

Click here for additional data file.

(A) Peritonitis duration; (B) History of GN. The information size was calculated based on relative risk reduction (RRR) or MD, alpha of 5%, power of 80%.

(TIF)

Click here for additional data file.

Results of sensitivity analysis.

(A) Age on dialysis. (B) D/P Cr. (C) PD duration. (D) peritonitis duration. (E) Use of icodextrin.

(TIF)

Click here for additional data file.

Checklist.

PRISMA 2020 checklist.

(DOCX)

Click here for additional data file.

Search strategy for the databases.

(DOCX)

Click here for additional data file.

The protocol of this review.

(PDF)

Click here for additional data file.

Risk assessment of the included studies using ROBINS-I tool.

(DOCX)

Click here for additional data file.

Publication bias associated with potential risk factors for EPS.

(DOCX)

Click here for additional data file.

5 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Additional Editor Comments:

The reviewers have raised a number of points which we believe major modifications are necessary to improve the manuscript, taking into account the reviewers' remarks. Please consider and address each of the comments raised by the reviewers before resubmitting the manuscript. This letter should not be construed as implying acceptance, as a revised version will be subject to re-review.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: No

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: 1. In the introduction (Line 97), risk factors for EPS have been mentioned which includes duration of PD, age at initiation, bio incompatible dialysate, number of peritonitis and duration, kidney transplantation.

Comment: Organ transplantation has been mentioned to increase the risk of EPS possibly related to use of immunosuppressive medications or withdrawal of PD (Fieren et al, Posttransplant encapsulating peritoneal sclerosis: a worrying new trend?).

In your meta-analysis, age at PD onset was significantly associated with development of EPS.

Given the number of young PD patients who undergo transplantation or have access to transplantation, I wonder if younger age at onset of PD and transplantation puts them at a higher risk of developing EPS in future. If data is available in the included case control studies, I would be curious to know if transplantation increased risk of EPS in your meta-analysis and consider addition of that to your manuscript.

2. Although young age has been implicated as a risk factor in your study and multiple studies before, is it confounded by the fact that younger patients may undergo PD for a longer duration of time, undergo transplantation and are alive long enough to develop EPS while older patients may not undergo PD for a longer duration and may not be alive long enough to develop EPS?

3. Use of Icodextrin was not a risk factor for development of EPS. I wonder if using Dianeal 2.5/4.25% is associated with higher risk of development of EPS as icodextrin may not be used for an entire dwell and since icodextrin is a water-soluble polymer may behave differently when compared to dextrose. Was there any information on other types of PD fluids used in the included studies?

Overall, I commend the authors for this well conducted meta-analysis to determine the risk factors associated with EPS.

Reviewer #2: I thank the authors and editors for this opportunity to review this manuscript.

My comments are as below:

1. The main weakness of the study lies in its eligibility criteria. We know that the aggregated evidence from observational studies is weaker than randomized controlled trials. In this regard, restricting the inclusion criteria to a particular type of observational study compromises the robustness and comprehensiveness of the review. Next, the study population remains poorly described. Moreover, the exclusion criteria didn't appear sound to me. The statement (sentence no. 130-132) of not recruiting poor-quality studies is more of a subjective decision of the authors and is not an endorsable way to eliminate studies.

2. The meta-analysis model choice guided by heterogeneity assessment is not recommended.

3. The cut-offs used to determine and quantify heterogeneity are not clear.

4. It's not advisable to use funnel plots for publication bias assessment when there are <10 studies available for inclusion in the meta-analysis models.

5. The abstract is not clear enough. I suggest a more simplified and logical presentation.

6. The total capitalization of some of the database names (e.g., PubMed) is not advisable.

7. The phrase "qualified studies" is better avoided.

8. Sentence 31-32 in the abstract is confusing. Ideally, co-stating the rationale of mentioning alternative model choices is required.

9. Sentence 44-45 in the abstract doesn't fit a conclusion statement. Better to avoid such in the abstract.

10. The study lacks a protocol to compare its pre-review notions to the presented methodology.

11. I couldn't find any mention of the supplementary tables and figures in the main manuscript text. These are simply listed before the reference section.

12. It's not clear how PubMed search strings got combined.

13. The claim of searching all relevant data as mentioned in sentence no. 116-117 is perhaps not appropriate.

14. The type of conflicting data authors is referring to in sentence no. 121-122 is not clear.

15. The tables lack publication-quality legends, footers, and formatting.

16. Sentence no. 150-152 is confusing.

17. Regarding sentence no. 157-159, without a study protocol, it's impossible to compare it with the preliminary plan made by the authors.

18. The PRISMA flow diagram is not adequately cited and presented.

19. The PRISMA checklist is not recommended anymore as a more recent version is available.

20. The figures lack proper legends. The requirements of different colors in forest plots are unclear.

21. The manuscript writing requires meticulous editing.

22. References use was inadequate.

Thank you.

Reviewer #3: Manuscript ID: PONE-D-21-39037

The authors present a comprehensive summary of their systematic review and meta-analysis of Risk factors for Encapsulating Peritoneal Sclerosis in patients undergoing peritoneal dialysis: a meta-analysis. They presented that lower age at PD onset, higher D/P Cr, longer PD duration, longer duration of peritonitis, and history of glomerulonephritis were risk factors for EPS. This review is interesting. However, there are some concerns that need to be addressed.

1. Grey literature: If the authors performed the search using specific gray literature, please provide the name of such database. Overall, I have concerns about the reproducibility of the meta-analysis. Certain aspects of the MOOSE guideline were followed, but there are some key flaws that may affect the summary effect estimate. Since search limited, unpublished studies were not included, I worry that this summary effect estimate will shrink or be nullified if grey literature, unpublished studies are found. The authors may only be able to say that their meta-analysis resolves discrepancies between published studies. I am hesitant because there are too many examples of unpublished data nullifying significant effects in meta-analyses. If the authors conduct a thorough search for unpublished data, this comment becomes irrelevant. Please add this information in supplementary.

2. This review was not followed PRISMA 2020 flow and absent of PRISMA 2020 checklist. The author should follow PRISMA 2020 flow diagram structure. In addition, the protocol registration was absent. Prospective registration of systematic reviews promotes transparency, helps reduce potential for bias and serves to avoid unintended duplication of reviews.

3. In "Methods" part, the independent investigators who extracted data and evaluated quality of each study should be clarify.

4. How were the search terms defined? Is there a pre-test to define the search strategy used in each database? Did the search strategy the same in all databases? I wonder if the search strategies have been developed with the help of a librarian or experienced reviewers in the field. Generally, if the search yield is too low, systematic reviewers would need to modify the search term to ensure that it well cover most of the related papers. This could be done by in the PICO such as Control and Outcome; employ a free text rather than thesaurus search terms. I think the keywords for the outcome domain were not comprehensive enough to capture all potential synonyms of the outcomes of interested in this study, and therefore would be best not to limit the search with these terms. An example of the search strategies used for a particular database would show a transparency in this step (I don’t think the keywords presented is sufficient as an example of a search strategies as recommended in PRISMA).

5. The date of the literature was April 1, 2021. In accordance with guidelines, the literature search should be performed until six months before the submission of the manuscript for publication. If the manuscript is accepted for publication, it is already outdated. In addition, the author should describe exact date and duration of the literature search in both abstract and main manuscript.

6. How authors dealt with missing data. Did you receive all answers from authors of the studies or make some imputation? In general, if the study did not report the data of the primary or secondary outcomes measures, the authors should contact via email to provide this information. Have you considered to contact authors via researchers’ network ResearchGate (https://www.researchgate.net/), Academia (https://www.academia.edu/), Loop (https://loop.frontiersin.org/) or Quora (https://es.quora.com/)? Nowadays these platforms are very useful and efficient canals to contact authors.

7. In quality assessment for observational study design part, I recommend the authors apply the ROBINS-I (Risk of Bias in Nonrandomized studies of Interventions) tool. The authors already applied the Newcastle Ottawa Scale, which is a validated tool and was an acceptable choice. However, to enhance the reproducibility and comparability of this review to future reviews of a similar topic (possibly an update of this review) I recommend including a risk of bias assessment using ROBINS-I, since it is the newest and most robust method of assessing risk of bias in systematic reviews/meta-analyses.

8. The authors should demonstrate both statistics and visualization. In addition, I suggest plot the funnel and contour-enhanced funnel in the graphic and not only the studies for better interpretation. Besides, it is necessary to present the p value for this analysis.

9. Finally, since I am not a native English user, I did not check for grammatical errors thoroughly. This should be done by an appropriate language reviewer.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Sumanta Saha

Reviewer #3: Yes: Wisit Kaewput

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

20 Feb 2022

Response to comment

We appreciate the time and efforts by the editor and referees in reviewing this manuscript. We have addressed all issues indicated in the review report, which may largely improve our current study.

Dear Dr. Yonggui Wu,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Feb 19 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols athttps://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

Answer: Thanks very much for your positive comments and kind reminding. We have revised the manuscript as suggested and re-submit our manuscript for fresh review. Meanwhile, This meta-analysis was registered with the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42022302786). The protocol of this review was presented in supplementary S3 File. Thank you again.

-------------------------------------

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

2. https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting sample_title_authors_affiliations.pdf

Answer: Thank you for your time in reviewing this manuscript. We have revised the format of this manuscript again based on PLOS ONE's style requirements, including file naming, tables formatting, and references formatting. Thank you again.

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Answer: Thanks very much for your suggestion. We have updated the ORCID iD for the corresponding author.

Reviewers' Comments to Author:

Reviewer: 1

Comments to the Author

Major comments

1. Organ transplantation has been mentioned to increase the risk of EPS possibly related to use of immunosuppressive medications or withdrawal of PD (Fieren et al, Posttransplant encapsulating peritoneal sclerosis: a worrying new trend?). In your meta-analysis, age at PD onset was significantly associated with development of EPS. Given the number of young PD patients who undergo transplantation or have access to transplantation, I wonder if younger age at onset of PD and transplantation puts them at a higher risk of developing EPS in future. If data is available in the included case control studies, I would be curious to know if transplantation increased risk of EPS in your meta-analysis and consider addition of that to your manuscript.

Answer: Thank you for arising this question. In some studies, a high incidence of post-transplantation EPS has been noted [ref. 1. Balasubramaniam, G, Nephrol Dial Transplant. 2009(PMID：19211652); 2. Afthentopoulos, I, Adv Ren Replace Ther. 1998(PMID：9686626)]. Balasubramaniam G, et al reported that the median time from renal transplantation to development of EPS was 5.4 months. As you said, The use of calcineurin inhibitor as a part of immunosuppressive regimens also increased the risk of EPS. For example, Tacrolimus and cyclosporine can promote fibrosis progression possibly by activating the epithelial-mesenchymal transition(EMT) response (increased TGF-β1 and α-SMA expression) and increasing the expression of fibrose-related genes [ref. 1. Liu QF, Int Urol Nephrol. 2017(PMID: 27796696); 2. Nam HK, Am J Nephrol. 2013(PMID: 23258196); 3. Lim BJ, Pediatr Nephrol. 2009(PMID：19066978); 4. Kern G, PLoS One. 2014(PMID：24816588)].

However, only one study reported some Variables related to kidney transplantation [Korte MR, Perit Dial Int. 2011(PMID：21454391)] between EPS group and NO EPS group, which demonstrated that a typical patient with EPS after kidney transplantation has started on PD at relatively young age (<50 years). Therefore, the meta-analysis for kidney transplantation was not performed here. Thank you again.

2. Although young age has been implicated as a risk factor in your study and multiple studies before, is it confounded by the fact that younger patients may undergo PD for a longer duration of time, undergo transplantation and are alive long enough to develop EPS while older patients may not undergo PD for a longer duration and may not be alive long enough to develop EPS?

Answer: Thank you for arising this question. Several studies have confirmed that the younger a patient is when at PD start, the greater the chance that EPS will develop. [ref. 1. Yamamoto R, Adv Perit Dial. 2002(PMID：12402604); 2. Johnson DW, Kidney Int. 2010(PMID：20375981)]. Given its in dependent significance, this variable is not simply a result of a longer time on PD or a longer duration of follow-up from PD initiation. As many articles have reported, they proposed the disrupted peritoneal fibrosis repair process in PD patients with younger age as possible cause of this observation [ref. 1. Korte MR, Perit Dial Int. 2011(PMID：21454391); 2. Chin AI, Am J Kidney Dis. 2006(PMID：16564950)].

In reaction to multiple insults with (for example) glucose, mesothelial cells secrete cytokines that subsequently lead to recruitment of macrophages and fibroblasts. Peritoneal fibrosis is then a result of disrupted repair, with fibrin deposition on a denudated mesothelial cell layer. Perhaps this repair process is more vigorous in younger patients, leading to earlier development of fibrosis. Therefore, we considered the age at PD onset as a risk factor for the development of EPS. We have added more explanations for this variable into the manuscript (Line 412-421, Page 23-24). Thank you again.

3. Use of Icodextrin was not a risk factor for development of EPS. I wonder if using Dianeal 2.5/4.25% is associated with higher risk of development of EPS as icodextrin may not be used for an entire dwell and since icodextrin is a water-soluble polymer may behave differently when compared to dextrose. Was there any information on other types of PD fluids used in the included studies?

Answer: Thank you for arising this question. As many articles have reported, PD fluids contain glucose, which is a major cause of peritoneal membrane injury. Heat sterilization of PD fluids also leads to the formation of glucose degradation products, which accelerate the formation of AGEs [ref. Moriishi, M, Adv Perit Dial. 2002(PMID：12402608)]. Moreover, patients who developed EPS had a higher cumulative glucose exposure [ref. 1. Lambie, M. L, Kidney Int. 2010(PMID：20571473); 2. Hendriks, P. M, Perit Dial Int. 1997(PMID：9159833)].

However, there was no reported data about other type of PD fluids used between EPS group and NO EPS group among studies. Thank you again.

Reviewer: 2

Comments to the Author

I thank the authors and editors for this opportunity to review this manuscript.

Answer:Thanks very much for your comments. We revised the manuscript as suggested.

Major comments

1. The main weakness of the study lies in its eligibility criteria. We know that the aggregated evidence from observational studies is weaker than randomized controlled trials. In this regard, restricting the inclusion criteria to a particular type of observational study compromises the robustness and comprehensiveness of the review. Next, the study population remains poorly described. Moreover, the exclusion criteria didn't appear sound to me. The statement (sentence no. 130-132) of not recruiting poor-quality studies is more of a subjective decision of the authors and is not an endorsable way to eliminate studies.

Answer: Thanks very much for your suggestion. We sincerely appreciate the valuable comment. Firstly, we re-searched relevant citations in five databases and one website from their inception to January 1st, 2022 without restriction by study type, and two prospective cohort studies included in this meta-analysis. Correspondingly, the contents of tables and figures in this study have been modified. Secondly, We have modified the content in Table 1, in which the basic characteristics of study population were further described (Page 12). Finally, We've revised our exclusion criteria in this study (Line 140-142, Page 7).

2. The meta-analysis model choice guided by heterogeneity assessment is not recommended.

Answer: Thanks very much for your suggestion. In PRISMA 2020, the fixed-effect model assumes that there is a common treatment effect for all included studies; it is assumed that the observed differences in results across studies reflect random variation. The random-effects model assumes that there is no common treatment effect for all included studies but rather that the variation of the effects across studies follows a particular distribution. However, differences in study design, interventions, and study populations of the included studies make it difficult to decide on the choice of meta-analysis model. Therefore, we intuitively selected a model based on the heterogeneity statistics observed. Thank you again.

3. The cut-offs used to determine and quantify heterogeneity are not clear.

Answer: Thank you for arising this question. In this meta-analysis, when I2 was > 50% and Q chisquared test result < 0.1, it shows that there is large heterogeneity among the trials. Once large heterogeneity was present, sensitivity analysis were performed to identify responsible outlier studies. Therefore, we deleted the Fig 5 and the results of sensitivity analysis were presented in S3 Fig. We have revised the content in the part of statistical analysis with red markers (Line 165-169, Page 9). Thank you again.

4. It's not advisable to use funnel plots for publication bias assessment when there are <10 studies available for inclusion in the meta-analysis models.

Answer: Thanks very much for your suggestion. Publication bias was evaluated using the Egger’s test in this study (Line 170, Page 9). The results of the Egger’s test for the analysis of EPS-related risk factors among PD patients were presented in S2 Table and showed that the included studies may have no possibility of publication bias. Thank you again.

5. The abstract is not clear enough. I suggest a more simplified and logical presentation.

Answer: Thanks very much for your suggestion. We removed unnecessary information, added the description of risk assessment and revised some data with red markers to make the abstract more simplified.

6. The total capitalization of some of the database names (e.g., PubMed) is not advisable.

Answer: Thanks very much for pointing out my mistakes. We have revised this error into the correct statement (Line 28, Page 2) (Line 120, page 6).

7. The phrase "qualified studies" is better avoided.

Answer: Thanks very much for your suggestion. We have revised this statement with red markers (Line 122-123, Page 6-7).

8. Sentence 31-32 in the abstract is confusing. Ideally, co-stating the rationale of mentioning alternative model choices is required.

Answer: Thanks very much for your suggestion. To further simplify the presentation of the abstract, we removed the sentence 31-32 and added the description of risk assessment in the part of methods with red markers (Line 31-33, Page 2).

9. Sentence 44-45 in the abstract doesn't fit a conclusion statement. Better to avoid such in the abstract.

Answer: Thanks very much for your suggestion. We removed sentence 44-45 and re-written the conclusion section of the abstract. Thank you again.

10. The study lacks a protocol to compare its pre-review notions to the presented methodology.

Answer: Thanks very much for your suggestion. As you suggested, this meta-analysis was registered at PROSPERO (CRD42022302786). The protocol of this review was presented in S3 File.

11. I couldn't find any mention of the supplementary tables and figures in the main manuscript text. These are simply listed before the reference section.

Answer: Thank you for arising this question. the supplementary tables and figures captions were listed after the reference section with red markers. Meanwhile, the contents of supporting information files were revised and uploaded separately in submission system. Thank you again.

12. It's not clear how PubMed search strings got combined.

Answer: Thank you for arising this question. We used a combination of search medical subject heading (MeSH) related to EPS and risk factor in PD patients. We made some changes to the search terms and the final strategies were peer reviewed by an experienced information specialist within our team. S2 File outlined the detailed search strategy of PubMed. Thank you again.

13. The claim of searching all relevant data as mentioned in sentence no. 116-117 is perhaps not appropriate.

Answer: Thanks very much for your suggestion. We have revised this statement with red markers(Line 120, Page 6).

14. The type of conflicting data authors is referring to in sentence no. 121-122 is not clear.

Answer: Thank you for arising this question. Conflicting data in sentence no. 121-122 refers to any discrepancies in study selection processes. We have revised this statement with red markers (Line 127, Page 7).

15. The tables lack publication-quality legends, footers, and formatting.

Answer: Thanks very much for your suggestion. We have revised format of the tables and made the necessary changes to the content of the tables. Legends and footnotes have been placed below the tables. Thank you again.

16. Sentence no. 150-152 is confusing.

Answer: Thank you for arising this question. In this study, continuous variables were summarized using mean difference (MD) or standardized mean difference (SMD) with their 95% confidence intervals (CI), for dichotomous variables, odds ratio (OR) with their 95% CI were estimated. We revised the sentence no. 150-152 in the section of statistical analysis with red markers (Line 161-164, Page 8).

17. Regarding sentence no. 157-159, without a study protocol, it's impossible to compare it with the preliminary plan made by the authors.

Answer: Thank you for arising this question. This is a typo. Because large heterogeneity was found in potential risk factors of exposure to age at PD onset, duration of PD, D/P Cr, peritonitis duration, and use of icodextrin, we wanted to perform subgroup analysis to explore sources of heterogeneity. However, we only performed sensitivity analysis due to the limited number of studies and data. We have deleted this inappropriate statement. Thank you again.

18. The PRISMA flow diagram is not adequately cited and presented.

Answer: Thanks very much for your suggestion. As you suggested, we have followed PRISMA 2020 flow diagram structure (Fig.1).

19. The PRISMA checklist is not recommended anymore as a more recent version is available.

Answer: Thanks very much for your suggestion. The PRISMA 2020 checklist has already uploaded again (S1 File). Thank you again.

20. The figures lack proper legends. The requirements of different colors in forest plots are unclear.

Answer: Thank you for arising this question. We added the proper descriptions in figures legends with red markers. In addition, different colors in forest plots represent different types of primitive data (e.g. green refers to Continuous variables, blue refers to dichotomous variables). The colors in these forest plots cannot be changed in the analysis software. Thank you again.

21. The manuscript writing requires meticulous editing.

Answer: Thanks very much for your suggestion. We have deleted some inappropriate presentations and polished some sentences as well the corresponding revision into manuscript with red markers in the sections of abstract, methods, results, discussion, and conclusion. Meanwhile, We invited native speakers to help us revise the grammar errors. Thank you again.

22. References use was inadequate.

Answer: Thanks very much for your suggestion. We have added more references for further verification in this study. Thank you again.

Reviewer: 3

Comments to the Author

The authors present a comprehensive summary of their systematic review and meta-analysis of Risk factors for Encapsulating Peritoneal Sclerosis in patients undergoing peritoneal dialysis: a meta-analysis. They presented that lower age at PD onset, higher D/P Cr, longer PD duration, longer duration of peritonitis, and history of glomerulonephritis were risk factors for EPS. This review is interesting. However, there are some concerns that need to be addressed.

Answer:Thanks very much for your comments. We revised the manuscript as suggested.

Major comments

1. Grey literature: If the authors performed the search using specific gray literature, please provide the name of such database. Overall, I have concerns about the reproducibility of the meta-analysis. Certain aspects of the MOOSE guideline were followed, but there are some key flaws that may affect the summary effect estimate. Since search limited, unpublished studies were not included, I worry that this summary effect estimate will shrink or be nullified if grey literature, unpublished studies are found. The authors may only be able to say that their meta-analysis resolves discrepancies between published studies. I am hesitant because there are too many examples of unpublished data nullifying significant effects in meta-analyses. If the authors conduct a thorough search for unpublished data, this comment becomes irrelevant. Please add this information in supplementary.

Answer: Thanks very much for your suggestion. We sincerely appreciate the valuable comment. We searched the http://www.greylit.org/ website for studies that were registered as completed but not yet published. However, no relevant studies were registered in this website. We have added it to the research strategy with red markers (Line 123-125, Page 7). Thank you again.

2. This review was not followed PRISMA 2020 flow and absent of PRISMA 2020 checklist. The author should follow PRISMA 2020 flow diagram structure. In addition, the protocol registration was absent. Prospective registration of systematic reviews promotes transparency, helps reduce potential for bias and serves to avoid unintended duplication of reviews.

Answer: Thanks very much for your suggestion. As you suggested, we have followed PRISMA 2020 flow diagram structure (Fig 1) and the PRISMA 2020 checklist has already uploaded again (S1 File). In addition, this meta-analysis was registered at PROSPERO (CRD42022302786). The protocol of this review was presented in S3 File. Thank you again.

3. In "Methods" part, the independent investigators who extracted data and evaluated quality of each study should be clarify.

Answer: Thanks very much for your suggestion. In this study, data extraction and quality assessment were completed by two authors (Yuanyuan Li, and Hanxu Zeng), which were detailed in the methods section (Line 129, Page 7) (Line 145, Line 150 Line 157, Page 8).

4. How were the search terms defined? Is there a pre-test to define the search strategy used in each database? Did the search strategy the same in all databases? I wonder if the search strategies have been developed with the help of a librarian or experienced reviewers in the field. Generally, if the search yield is too low, systematic reviewers would need to modify the search term to ensure that it well cover most of the related papers. This could be done by in the PICO such as Control and Outcome; employ a free text rather than thesaurus search terms. I think the keywords for the outcome domain were not comprehensive enough to capture all potential synonyms of the outcomes of interested in this study, and therefore would be best not to limit the search with these terms. An example of the search strategies used for a particular database would show a transparency in this step (I don’t think the keywords presented is sufficient as an example of a search strategies as recommended in PRISMA).

Answer: Thank you for arising this question. We attached great importance to this comment. As you proposed, we redefined the search terms based on a PICO-style approach with the help of a reviewer. Firstly, five known relevant studies were used to identify records within databases. Secondly, Candidate search terms were identified by looking at words in the titles, abstracts and subject indexing of those records. A draft search strategy was developed using those terms and additional search terms were identified from the results of that strategy. Search terms were also identified and checked using the PubMed PubReMiner word frequency analysis tool. Finally, The strategy was developed by an information specialist and the final strategies were peer reviewed by an experienced information specialist within our team. The Search terms in each database were the same, while search strategy was different. S2 File outlined the detailed search strategy of PubMed. Thank you again.

5. The date of the literature was April 1, 2021. In accordance with guidelines, the literature search should be performed until six months before the submission of the manuscript for publication. If the manuscript is accepted for publication, it is already outdated. In addition, the author should describe exact date and duration of the literature search in both abstract and main manuscript.

Answer: Thanks very much for your valuable suggestion. We re-searched relevant citations in five databases and one website from their inception to January 1st, 2022, which were detailed in the sections of abstract and methods and made some changes to the search terms. A total of 495 potentially relevant articles were retrieved initially. Finally, we included 10 standards-compliant articles. Correspondingly, the contents of tables and figures in this study have been modified. Thank you again.

6. How authors dealt with missing data. Did you receive all answers from authors of the studies or make some imputation? In general, if the study did not report the data of the primary or secondary outcomes measures, the authors should contact via email to provide this information. Have you considered to contact authors via researchers’ network ResearchGate (https://www.researchgate.net/), Academia (https://www.academia.edu/), Loop (https://loop.frontiersin.org/) or Quora (https://es.quora.com/)? Nowadays these platforms are very useful and efficient canals to contact authors.

Answer: Thank you for arising this question. We were lucky that in this study, there was no data missing in the extraction process.

7. In quality assessment for observational study design part, I recommend the authors apply the ROBINS-I (Risk of Bias in Nonrandomized studies of Interventions) tool. The authors already applied the Newcastle Ottawa Scale, which is a validated tool and was an acceptable choice. However, to enhance the reproducibility and comparability of this review to future reviews of a similar topic (possibly an update of this review) I recommend including a risk of bias assessment using ROBINS-I, since it is the newest and most robust method of assessing risk of bias in systematic reviews/meta-analyses.

Answer: Thanks very much for your valuable suggestion. As you suggested, potential bias was evaluated by the ROBINS-I tool for observational studies. Finally, all studies presented a moderate overall risk of bias, in which confounding bias remains in all studies due to their non-randomized nature. The exact details of the risk evaluation are summarized in S1 Table. Thank you again.

8. The authors should demonstrate both statistics and visualization. In addition, I suggest plot the funnel and contour-enhanced funnel in the graphic and not only the studies for better interpretation. Besides, it is necessary to present the p value for this analysis.

Answer: Thanks very much for your suggestion. However, one reviewer suggested that the use of funnel plots for publication bias assessment is inappropriate in this study when there are ≤10 studies available for inclusion in the meta-analysis models. Therefore, publication bias of included studies was evaluated again using the Egger’s test. The results of Egger’s test were presented in S2 Table. Thank you again.

9. Finally, since I am not a native English user, I did not check for grammatical errors thoroughly. This should be done by an appropriate language reviewer.

Answer: Thanks for your time in reviewing this manuscript. We have invited native speakers to help us revise the grammar errors.

We tried our best to improve the manuscript and made some changes in the manuscript. These changes will not influence the content and framework of the paper. And here we did not list the changes but marked in red in revised paper.

We appreciate for Editors/Reviewers’ warm work earnestly, and hope that the correction will meet with approval. Once again, thank you very much for your comments and suggestions.

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

7 Mar 2022

Risk factors for Encapsulating Peritoneal Sclerosis in patients undergoing peritoneal dialysis: a meta-analysis

PONE-D-21-39037R1

Dear Dr. wu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

Additional Editor Comments:

It appears that all comments have been appropriately responded to. I have no further comments and recommend publication.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: No further comments to the authors. Authors have addressed all my comments. Manuscript is well written.

Reviewer #3: Thank you. I am satisfied with the authors replies and modifications and have no further suggestions.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: Yes: Wisit Kaewput

10 Mar 2022

PONE-D-21-39037R1

Risk factors for Encapsulating Peritoneal Sclerosis in patients undergoing peritoneal dialysis: a meta-analysis

Dear Dr. Wu:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Wisit Cheungpasitporn

Academic Editor

PLOS ONE