Every year, an increasing number of patients with end-stage renal disease (ESRD) are treated with hemodialysis.1 In regular hemodialysis, patients may be exposed to about 400 L of water per week across the semipermeable membrane of the hemodialysis or hemodiafilter.2 It is clear that water quality can have a significant impact on patient outcomes, so clean water used in the preparation of dialysis fluid is an absolute requirement for adequate hemodialysis and related therapies.3 Renal services in hospitals frequently derive their water supply from the municipal water distribution network.4 Municipal water needs additional purification to control chemical and biological contamination in dialysis centers.5 Contamination of hemodialysis fluid with whole cell of bacteria or their fragments is a significant health risk for hemodialysis patients with stimulating leukocytes to produce proinflammatory cytokines.6
The European Renal Associations recommends values of ≤100 colony-forming units (CFU)/ml of viable bacteria and ≤0.5 IU/ml endotoxin as safety criteria for hemodialysis fluid.7 Moreover, the American Association for the Advancement of Medical Instrumentation (AAMI) suggests allowable levels of ≤100 CFU/ml for viable bacteria and ≤0.25 IU/ml for endotoxins in hemodialysis fluid.8
Allowable levels of viable bacteria of 200 CFU/ml and less than 2 IU/ml of endotoxins in water used to prepare dialysate are recommended by AAMI.8 The corresponding value is 2,000 CFU/ml for the final dialysis fluid.8
Most bacteria are able to survive in biofilm and multiply within different parts of the hemodialysis water supply system by using trace amounts of organic and inorganic nutrients.9 This presents a potentially serious problem for patients.10
Bacterial enumeration by plating on standard growth media (Mueller-Hinton or Blood agar) and simple identification by biochemical testing is the only microbiological method for quality control of hemodialysis water in hospitals in Iran.
Monitoring and identification of bacterial communities at different points of hemodialysis system is of critical importance. The aim of this study was to evaluate the bacterial communities in hemodialysis water at different points of one educational hospitals in Tehran, and four in Alborz provinces.
Water samples were collected at the dialysis centers of one hospital in Tehran and four hospitals in Alborz cities from February 2014 to May 2014. Five points were selected for the collection of samples based on the dialysis systems used in each hospital (Figure 1). Two hospitals did not have municipal water reservoir stocks; in these municipalities, tap water was used for sampling. Sufficient water from municipal reservoirs was usually provided to hospitals to ensure water availability 24 hours a day, and was located in same place (adjacent to the pretreatment section of the hemodialysis treatment system). Over 4 months, a total of 80 samples were collected from all five hospitals from five points (sampling was repeated from the same points each month). The samples were collected from the following points: Municipal tap water (point 1); municipality water reservoir stock (point 2); the outlet port of the water-treatment system (point 3); reverse osmosis (RO; point 4); and fresh dialysate (before or after patient dialysis; point 5).11 Samples were taken 1 day before the routine monthly disinfection, which uses peracetic acid. The water samples were aseptically obtained in sterile bottles and immediately transported in cool conditions (10°C) to the Microbiology Research Laboratory at the School of Medicine, Alborz University of Medical Science, Alborz, Iran.
To estimate the number of live heterotrophic bacteria suspended in each undiluted water sample, 0.1 and 0.5 ml of each sample was processed on Reasoner’s 2A agar (R2A; Merck, Germany) using the spread plate technique. The plates were incubated at 20 ± 2°C for 7 days. This method is based on ISO 13959:2014, which provides guidelines and protocols for the monitoring of water for hemodialysis and related therapies.8
Routine Quality Control of Hemodialysis Water
The hospitals adapted the conventional urine culture methods to culture hemodialysis water. The hospitals cultured 10 μl of water samples on Blood agar and or Mueller-Hinton agar and incubated the culture media overnight at 37°C. The isolates were identified to the species level using conventional biochemical tests.
Viable Bacterial Counts
In this study, after an incubation time of 7 days, the colonies were counted on R2A agar plate (0.5 and 0.1 ml of inoculated samples), from 0 to 100 per plate, and multiplied by 2 and 10 to give the amounts of CFU/ml, respectively. Plates with more colonies were not counted and reported as 1,000 < CFU/ml. On the basis of colony morphology, pigment, gram staining, and a cytochrome-oxidase test, the organisms were classified into different groups.12
Molecular Identification Using 16S rRNA Sequencing
Genomic DNA was extracted from bacterial cultures using a DNA isolation kit (Qiagen, Valencia, CA) according to the manufacturer´s instructions and stored at −20°C until use.
Nearly the full length of the 16S rRNA genes from the isolates were amplified using primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1525R (5′-TGCACACAGGCCACAAGGGA-3′) as described previously.13 The sequences of amplified PCR products of the 16S rRNA gene for each isolate were determined using an ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Foster City, CA) and by following the supplier’s standard protocol.
Analysis of Sequence Data
The obtained sequences for each isolate from 16S rRNA was aligned separately and compared with all existing relevant sequences of bacteria retrieved from the GenBank database. Phylogenetic trees were obtained from DNA sequences using the neighbor joining (NJ) method Kimera 2-parameter (K2P) distance correction model with 1,000 bootstrap replications supported by the MEGA 4.1 software.14
Routine Quality Control of Hemodialysis Water by Hospitals
The hospital located in Tehran reported the isolation of gram-negative bacteria from hemodialysis fluid that was identified as Pseudomonas aeruginosa. All the hospitals in Alborz reported no growth from the samples obtained from their hemodialysis centers after an 18 hour incubation time.
Colony Counts of Viable Bacteria
In this study, high cell counts of bacteria were recorded on R2A medium plates from the samples of one hospital at Tehran and four hospitals in the Alborz province. In the case of the hospital in Tehran, only one the outlet ports of dialysis fluid drains showed high amounts of culturable bacteria (≥103). The majority of samples from the four hospitals in Albourz province had colony counts ≥1,000 CFU/ml at points 2, 3, and 4 (Table 1). At all points, the counts in the hemodialysis fluid were higher than the maximum recommended values.8 The numbers of culturable bacteria gradually increased as the water flowed from the municipal water reservoir to the outlet ports of dialysis fluid in all the hospitals studied (Table 1). One hospital had contamination of 5.7 CFU/ml at point 1, but no contamination at points 2 and 3, and 3.5 CFU/ml at point 4. At this setting, 251 CFU/ml bacteria were recovered at point 5. For the other hospitals, heavy contamination with culturable bacteria was detected (Table 1).
According to hospital reports on counts of Blood agar or Mueller-Hinton agar, viable bacteria were negative for all samples, except for one report of P. aeruginosa from point 3 of the hospital in Tehran.
In this study, based on primary grouping of the isolates by conventional microbiological tests, a total of 229 morphotypes were recovered from the samples from all hospitals. These were grouped into 45 clusters.
Out of 80 samples from points at all hemodialysis centers, a total of 229 isolates were studied. According to phenotypic characteristics, the isolates were classified into 45 groups, of which 28 were Gram positive and 17 groups were Gram negative. From each group, one isolate was randomly selected and further studied using 16S rRNA gene sequencing. Out of 45 isolates, 28 isolates were identified as Gram-positive bacteria, including 20 isolates clearly assigned to 16 valid previously published taxa (Figure 2).
Isolate numbers DW20 and DW814 showed the highest similarity (99.9%) to that of Kocuria rosea and formed a single line in the phylogenetic tree (Figure 2). DW200 was identified as Arthrobacter globiformis, with 99.8% similarity. The isolates DW248 and DW 258 showed the highest similarity (99.9%) to that of Arthrobacter agilis, and the isolate DW413 was also identified as Arthrobacter agilis with 99.5% similarity (seven nucleotide mismatches). DW818 was identified as Micrococcus luteus (99.8% similarity) and the isolates DW 645, DW664, and DW 665 showed identical 16s RNA and were assigned as Microbacterium testaceum (99.9% similarity). Isolates DW414, DW821, DW239, DW204, and DW3 were, respectively, identified as Agrococcus jenensi, Dietzia schimae, Paenisporosarcina indica, Bacillus pumilus, and Staphylococcus pasteuri because of the high similarities of their 16S rRNA to that found in these bacteria.
The isolate DW233 showed 99.6% similarity to Staphylococcus xylosus, and was 99.4% similar to Staphylococcus saprophyticus. Because of their 16S rRNA, the isolates DW409, DW659, DW408, and DW660 were identified as Mycobacterium chelonae, Mycobacterium canariasense, Mycobacterium fluoranthenivorans, and Mycobacterium sacrum.
Eight isolates could not be identified to the species level, and their identity remains unknown because of the low similarity of their 16S rRNA gene sequences that found in valid species. The isolate DW458 sp showed 99.2% similarity (10 nucleotide differences) to that of Mycobacterium canariasense and formed a distinct line of phylogeny among the mycobacterium cluster. The isolate DW1 sp showed 99.2% similarity (11 nucleotide differences) to that of Kocuria polaris. The isolate DW602 sp showed 96.8% similarity (43 nucleotide mismatches) to Nesterenkonia lacusekhoensis, whereas isolates DW209 sp and DW 229 sp were 99% (12 nucleotide mismatches) similar to the Nesterenkonia lacusekhoensis.
DW221 sp showed 97.3% (36 nucleotide mismatches) similarity to Arthrobacter agilis and DW628 sp showed 98% similarity (23 nucleotide differences) to Microbacterium trichotecenolyticum and formed a distinct line in the phylogenic tree (Figure 2).
Out of 45 isolates, 17 isolates were Gram negative. Of these, 15 isolates were clearly assigned a species level by means of 16S rRNA. Two isolates remained unidentifiable because of their low similarity by 16S rRNA (Figure 3). The isolates DW206, DW224, and DW238 were identified as isolates Acinetobacter lwoffii (99.8% similarity), and DW4 was assigned to species level as Acinetobacter baumannii (99.5% similarity). The isolates DW817, DW813 and DW647, DW424, DW405 and DW405, DW605, DW666, DW827, DW830, and DW651 were identified as Psychrobacter pulmonis, Halomonas stevensii, Pseudomonas stutzeri, Pseudomonas mendocina, Mesorhizobium amorphae, Paracoccus aestuarii, Burkholderia cepacia, Herbaspirillum huttiense, and Pigmentiphaga kullae, respectively.
The isolate DW649 formed a distinct line in the phylogenetic tree and was similar to the 16S rRNA sequences of uncultured Sphingomonas sp. (Genbank: GU560166). It is clear that this isolate belongs to the Sphingomonas genus; however, the identity of this isolate remains unclear.
Based on 16S rRNA, the isolate DW653 sp showed 99.2% similarity (12 nucleotide mismatches) to Hydrogenophaga palleronii. The majority of the isolates belong to the genus Acinetobacter, Pseudomonas, Mycobacterium, and Staphylococcus, recovered from point 5.
Water used during dialysis sessions may be responsible for the transmission of an increasing number of infections, mainly caused by the presence of bacteria, endotoxins, and bacteria-derived products.15 Standard protocols are available to address the quality of hemodialysis water; however, because of a lack of knowledge and awareness, a simple approach consisting of culturing of 10 μl of water samples on Blood agar and or Mueller-Hinton agar for the enumeration of bacteria and identification of recovered isolates by conventional biochemical testing is routinely applied in our hospitals. We investigated the bacteriological quality of the main water of the hemodialysis centers in our setting according to standard protocols7,8,12,16 to assess contamination in treated water to evaluate the quality of hemodialysis water.
The municipal water sample in one hospital from Tehran showed 5.7 CFU/ml, whereas for one hospital at Alborz, 757.5 CFU/ml was counted (Table 1). In this study, hemodialysis water from four hospitals in Alborz showed the high levels of contamination (Table 1). Three other hospitals in Alborz have reservoir systems in the hospital that connect to the municipal water system. These hospitals also had 15.7, 254.5, and 259 CFU/ml of culturable bacteria in their reservoir systems. Coliforms were not detected in the municipal water systems of Tehran or Alborz in our study. The microbial standards for municipal drinking water define ≤500 CFU/ml of heterotrophic bacteria17 and no growth of total Coliforms (including fecal coliform and Escherichia coli).18 In other words, the microbiological quality of municipal water that flows to hemodialysis centers in Tehran and Alborz met drinking water regulations.
The AAMI recommends culturing hemodialysis fluids on Trypticase soy agar (TSA) at 37°C for 48 hours.8 However, it has been reported that R2A medium in combination with low temperature (25 ± 2°C) and an extended incubation time (>7 days), apparently improved the isolation and monitoring of culturable microorganisms in hemodialysis water.12,19 Routinely, in our hospital settings, culturing of only 10 μl of water samples on Blood agar and or Muller-Hinton agar have been used, which led to either no growth or the isolation of only a few commonly recovered bacteria (such as P. aeruginosa, at one hospital at Tehran). Isolation of P. aeruginosa has been reported by Ekrami et al.20 from the water supply in a hemodialysis ward, whereas no isolation of bacteria was reported by Alizadeh et al.21 This underscores the need for changes in the monitoring of culturable bacteria in Iran. It has been reported that using R2A culture media leads to significantly higher bacterial recovery than TSA media for both dialysis water and dialysates.12 In the current study, for most of the samples, Blood agar or Muller-Hinton agar produced negative results for contaminant bacteria. Meanwhile, culturing of samples on R2A media with a 7 day incubation time at 20°C markedly improved the recovery of culturable bacteria (Table 1), which confirms that using R2A media instead of other culture media is essential for accurate monitoring of bacterial contamination in hemodialysis systems.
In this study, for the first time, a standardized protocol for isolation8,12,17 and identification by means of 16S rRNA sequencing was used.13 One hospital in Tehran showed negative for culturable bacteria at points 2 and 3, but was positive at point 4 (3.5 CFU/ml) and point 5 (251 CFU/ml). However, at different points in the Alborz four hospitals, culturable bacteria CFU/ml increased from point 2 (225–259 CFU/ml) to point 5 (507–1000 CFU/ml). In all circumstances in the four Alborz hospitals, the counts in the hemodialysis fluid at different points were significantly higher than the AAMI’s maximum recommended values.8,12
In line with our findings, high levels of contamination in dialysis water fluid were reported in different medical settings. In a study conducted in the USA 25 years ago, 53% of dialysis centers had bacterial counts above the AAMI standard of ≤100 CFU/ml, 200 CFU/ml for water, and 35% of the centers had bacterial counts above the 2,000 CFU/ml standard for dialysate in at least one sampling period.22
At hemodialysis centers in nine hospitals in Japan, Oie et al.,16 reported high contamination of dialysate. Of 40 dialysate samples analyzed, 42.5% showed a bacterial count of more than 2,000 CFU/ml, which was above the AAMI standard. In Germany, dialysate of 30 dialysis centers was examined, and in 11.7% of all dialysate samples, contamination higher than the recommendations for dialysate (of 2,000 CFU/ml) were found. In 12.2% of all water sampled and 27.5% of all dialysate samples, values of 5 endotoxin units/ml were found.15
Polyvinylchloride (PVC) is a common substratum used in hemodialysis systems in all the investigated hospitals in our setting, and the hemodialysis systems in each hospital were disinfected using their protocols (no data available). Loose connections and multiple 90° turns in all hemodialysis systems were identified through direct observation and recording. In addition, most of the water distribution systems ranged in age from 8 to 22 years old (Table 1). Cappelli et al.23 have noted that PVC alterations, induced by time, could support bacterial proliferation and thus reduce the quality of dialysate. Andrysiak et al.24 have recommended using proper angles to permit optimum flow velocity. We speculate that changes in PVC over time, aging distribution systems and the presence of multiple 90° turns might be contributing factors, as these can all create surfaces that support bacterial growth and biofilm formation in hemodialysis water systems.
Different genera among Gram-positive bacteria were identified, including Arthrobacter, Agrococcus, Bacillus, Dietzia, Kocuria, Micrococcus, Microbacterium, Mycobacterium, Paenisporosarcina, Staphylococcus and some unknown species in the genera Arthrobacter, Kocuria, Microbacterium, Mycobacterium, and Nesterenkonia (Figure 1). Gram-negative bacteria were also identified, including Acinetobacter, Burkholderia, Halomonas, Herbaspirillum, Mesorhizobium, Paracoccus, Pigmentiphaga, Pseudomonas, and Psychrobacter. Some isolates among Gram-negative bacteria could not be clearly identified to the species level and were grouped between the Hydrogenophaga and Sphingomonas genera (Figure 2). Diverse bacterial communities consisting of different species of heterotrophic bacteria were reported in different hemodialysis settings.12,13,15,25,26 Our data revealed that the numbers of culturable bacteria gradually increased as water flowed from the municipality’s water reservoir to the outlet ports for dialysis fluid in all studied hospitals. These findings suggest that the dialysis systems and tubing along the fluid pathways within dialysis supplies are the main sources of contamination and biofilm development, and result in the high levels of bacterial contamination at different sampling points.
Some of the isolated bacteria, such as Sphingomonas and P. aeruginosa, are able to attach to the inner surface of tubing or filters by the production of slime.25,27 Some isolates belonging to mycobacteria also were recovered in our study; these have hydrophobic surfaces, facilitating the formation of biofilms.26,28 Some species isolated in our study can use traces of organic and inorganic nutrients in water and can also form biofilms.27,29 Recovery of Gram-positive bacteria such as Staphylococcus and Bacillus in our study might be attributed to, as Gomila et al.13 suggest, the maturation of biofilm in hemodialysis systems. Our results might provide evidence of biofilm formation in hemodialysis systems, which acts as a reservoir for continuous contamination of hemodialysis systems28–31; however, further study is necessary to determine the role of recovered bacteria in biofilm formation and planktonic bacteria in hemodialysis water systems.
Bacterial growth and biofouling within hemodialysis systems can cause serious problems for patients because of the presence of bacteria or their byproducts in purified water.30–33 In our setting, the impact of such contamination for hemodialysis patients is not determined. However, taxa such as Acinetobacter, Burkholderia Pseudomonas, Mycobacterium, and Staphylococcus have been described previously as causative agents of infections or related to pyrogenic reactions during dialysis sessions.32–38
From a laboratory point of view, standard protocols are available to address the quality of hemodialysis water.7,8,36,37,39 However, in our setting, current monitoring of bacterial communities in hemodialysis water is usually restricted to the reporting of total viable count or the detection of specific groups of bacteria (Enterobacteriaceae and Pseudomonas) using culturing of water samples on Blood agar or Mueller-Hinton agar with a short incubation time (16–24 hours) at 37°C. It seems that poorly designed water treatment systems (multiple 90° turns), use of PVC for a long time (8–22 years, Table 1), and lack of proper maintenance are other problems in our settings that we must tackle to control microbial contamination of hemodialysis water.
In conclusion, because of the susceptibility of hemodialysis systems to microbial contamination and biofilm formation, it is necessary to take measures such as regular monitoring, and identification of culturable bacteria by applying standard protocols for evaluation of microbial and chemical contamination. Further studies are mandatory to address the impacts of diverse bacterial communities in hemodialysis systems and reduce risks to patient health.
The authors appreciate the financial support of Alborz University of Medical Sciences, Alborz, Iran (Grant No. 2247746).
1. Makusidi MA, Liman HM, Yakubu A, Isah MD, Abdullahi S, Chijioke A: Hemodialysis
performance and outcomes among end stage renal disease patients from Sokoto, North-Western Nigeria. Indian J Nephrol 2014.24: 8285.
2. Pontoriero G, Pozzoni P, Andrulli S, Locatelli F: The quality of dialysis water
. Nephrol Dial Transplant 2003.18(Suppl 7): vii215; discussion vii56.
3. Ward RA: Dialysis water
as a determinant of the adequacy of dialysis. Semin Nephrol 2005.25: 102111.
4. Hoenich NA: Disinfection of the hospital water
supply: A hidden risk to dialysis patients. Crit Care 2009.13: 1007.
5. Vorbeck-Meister I, Sommer R, Vorbeck F, Hörl WH: Quality of water
used for haemodialysis: Bacteriological and chemical parameters. Nephrol Dial Transplant 1999.14: 666675.
6. Glorieux G, Schepers E, Schindler R, et al.: A novel bio-assay increases the detection yield of microbiological impurity of dialysis fluid, in comparison to the LAL-test. Nephrol Dial Transplant 2009.24: 548554.
7. Group EBPGW: European best practice guidelines for haemodialysis (part 1). Nephrol Dial Transplant 2002.17: 1111.
8. ISO 13959: Water
for haemodialysis and related therapies, 2014.
9. Kulakov LA, McAlister MB, Ogden KL, Larkin MJ, O’Hanlon JF: Analysis of bacteria contaminating ultrapure water
in industrial systems. Appl Environ Microbiol 2002.68: 15481555.
10. Jochimsen EM, Carmichael WW, An JS, et al.: Liver failure and death after exposure to microcystins at a hemodialysis
center in Brazil. N Engl J Med 1998.338: 873878.
11. Montanari LB, Sartori FG, Cardoso MJ, et al.: Microbiological contamination of a hemodialysis
distribution system. Rev Inst Med Trop Sao Paulo 2009.51: 3743.
12. van der Linde K, Lim BT, Rondeel JM, Antonissen LP, de Jong GM: Improved bacteriological surveillance of haemodialysis fluids: A comparison between Tryptic soy agar and Reasoner’s 2A media. Nephrol Dial Transplant 1999.14: 24332437.
13. Gomila M, Gascó J, Busquets A, et al.: Identification of culturable bacteria present in haemodialysis water
and fluid. FEMS Microbiol Ecol 2005.52: 101114.
14. Tamura K, Dudley J, Nei M, Kumar S: MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 2007.24: 15961599.
15. Bambauer R, Schauer M, Jung WK, Daum V, Vienken J: Contamination of dialysis water
and dialysate. A survey of 30 centers. ASAIO J 1994.40: 10121016.
16. Oie S, Kamiya A, Yoneda I, et al.: Microbial contamination of dialysate and its prevention in haemodialysis units. J Hosp Infect 2003.54: 115119.
17. Coulliette AD, Arduino MJ: Hemodialysis
quality. Semin Dial 2013.26: 427438.
18. WHO G: Guidelines for drinking-water
19. Punakabutra N, Nunthapisud P, Pisitkun T, Tiranathanagul K, Tungsanga K, Eiam-Ong S: Comparison of different culture methods on bacterial recovery in hemodialysis
fluids. J Med Assoc Thai 2004.87: 13611367.
20. Ekrami A, Kayedani A, Jahangir M, Kalantar E, Jalali M: Isolation of common aerobic bacterial pathogens from the environment of seven hospitals, Ahvaz, Iran. JJM 2011.4: 7582.
21. Alizadeh M, Bazrafshan E, Jafari MH, Razabzadeh A: Microbiological and chemical indicators of water
used in hemodialysis
centers of hospitals affliated to Zahedan University of medical science, 2012. Health Dev 2013.3: 182190.
22. Harding GB, Klein E, Pass T, Wright R, Million C: Endotoxin and bacterial contamination of dialysis center water
and dialysate; a cross sectional survey. Int J Artif Organs 1990.13: 3943.
23. Cappelli G, Ballestri M, Facchini F, Carletti P, Lusvarghi E: Leaching and corrosion of polyvinyl chloride (PVC) tubes in a dialysis water
distribution system. Int J Artif Organs 1995.18: 261263.
24. Andrysiak P, Varughese PM: Design requirements for a water
distribution system in a hemodialysis
center. Dial Transplant 2002.31: 683691.
25. Rosenberg J: Primary bloodstream infections associated with dialyzer reuse in california dialysis centers: 43rd Annual Meeting, Idsa, 2005.
26. Gomila M, Ramirez A, Lalucat J: Diversity of environmental Mycobacterium isolates from hemodialysis water
as shown by a multigene sequencing approach. Appl Environ Microbiol 2007.73: 37873797.
27. Stoodley P, Sauer K, Davies DG, Costerton JW: Biofilms as complex differentiated communities. Annu Rev Microbiol 2002.56: 187209.
28. Rosenberg J. Primary blood stream infections associated with dialyzer reuse in California dialysis centers.Proceeding of the 43rd Annual Meeting of the Infectious Diseases Society of America. 2005.IDSA, Alexandria, VA.
29. Pasmore M, Marion K: Shirtliff M, Leid JG, Biofilms in hemodialysis
. The Role of Biofilms in Device-Related Infections. 2009, New York, NY, Springer, pp. 167192.
30. Man NK, Degremont A, Darbord JC, Collet M, Vaillant P: Evidence of bacterial biofilm in tubing from hydraulic pathway of hemodialysis
system. Artif Organs 1998.22: 596600.
31. Dasgupta MK: Biofilms and infection in dialysis patients. Semin Dial 2002.15: 338346.
32. Berkelman RL, Godley J, Weber JA, et al.: Pseudomonas cepacia peritonitis associated with contamination of automatic peritoneal dialysis machines. Ann Intern Med 1982.96: 456458.
33. Welbel SF, Schoendorf K, Bland LA, et al.: An outbreak of gram-negative bloodstream infections in chronic hemodialysis
patients. Am J Nephrol 1995.15: 14.
34. Bolan G, Reingold AL, Carson LA, et al.: Infections with Mycobacterium chelonei in patients receiving dialysis and using processed hemodialyzers. J Infect Dis 1985.152: 10131019.
35. Flaherty JP, Garcia-Houchins S, Chudy R, Arnow PM: An outbreak of gram-negative bacteremia traced to contaminated O-rings in reprocessed dialyzers. Ann Intern Med 1993.119: 10721078.
36. Lowry PW, Beck-Sague CM, Bland LA, et al.: Mycobacterium chelonae infection among patients receiving high-flux dialysis in a hemodialysis
clinic in California. J Infect Dis 1990.161: 8590.
37. Hellou E, Artul S, Omari S, Taha M, Armaly Z, Nseir W: Non catheter-related bacteremia caused by Pseudomonas oryzihabitans in a patient undergoing hemodialysis
. Hemodial Int 2014.18: 711713.
38. Loo LW, Liew YX, Choong HL, Tan AL, Chlebicki P: Microbiology
and audit of vascular access-associated bloodstream infections in multi-ethnic Asian hemodialysis
patients in a tertiary hospital. Infect Dis (Lond) 2015.47: 225230.
39. Pérez García R, González Parra E, Ceballos F, Escallada Cotero R, Gómez-Reino MI, Martín-Rabadán P: Guía de gestión de calidad del líquido de diálisis. Nefrología 2004.24: 142.