Indian Journal of Nephrology About us |  Subscription |  e-Alerts  | Feedback | Login   
  Print this page Email this page   Small font sizeDefault font sizeIncrease font size
 Home | Current Issue | Archives| Ahead of print | Search |Instructions |  Editorial Board  

Users Online:1182

Official publication of the Indian Society of Nephrology
 ~  Similar in PUBMED
 ~  Article in PDF (489 KB)
 ~  Citation Manager
 ~  Access Statistics
 ~  Reader Comments
 ~  Email Alert *
 ~  Add to My List *
* Registration required (free)  

  Chemical Purity
   Microbiological ...
  Endotoxin Levels
   Article Figures
   Article Tables

 Article Access Statistics
    PDF Downloaded266    
    Comments [Add]    

Recommend this journal


  Table of Contents  
Year : 2020  |  Volume : 30  |  Issue : 7  |  Page : 18-23

Water treatment

Date of Web Publication15-Jul-2020

Correspondence Address:
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0971-4065.289827

Rights and Permissions

How to cite this article:
. Water treatment. Indian J Nephrol 2020;30, Suppl S1:18-23

How to cite this URL:
. Water treatment. Indian J Nephrol [serial online] 2020 [cited 2021 Jul 23];30, Suppl S1:18-23. Available from:

The average HD patient is exposed to approximately 25 times the amount of water normally ingested by an individual in a day. Exposure is without the protective barrier of the gastrointestinal (GI) tract and the detoxification function of the kidneys, increasing the risk of toxicity caused by the numerous chemical and microbiological contaminants in the water. The final quality of the water is dependent on the configuration of the treatment system and the quality of the feed water. High-flux dialyzers and HDF demand the use of ultrapure water. Dialysis units should design a system capable of generating appropriate quality water.

We recommend that all HD units should have water treatment systems designed to achieve the water quality of the Association for the Advancement of Medical Instrumentation (AAMI) standards [Table 1].
Table 1: Comparison of maximum permissible water contaminant levels and methods of analysis recommended by the European Pharmacopoeia and the Association for the Advancement of Medical Instrumentation

Click here to view

We suggest that the HD units should aim to achieve the European Standards of purity of water [Table 2].
Table 2: Maximum levels of the different water purity grades

Click here to view

We recommend that treated water should meet with the following three standards of water quality:

  1. Chemical purity
  2. Microbiological purity
  3. Endotoxin purity.

We recommend that a water treatment system should consist of the following:

  1. Pretreatment of the source water
  2. Reverse osmosis (RO)
  3. Storage facilities
  4. Distribution of treated water.

We suggest that additional chemical purity can be obtained by the addition of a deionizer, either a mixed bed or a cation and anion exchanger. It is important that the resins should be regularly regenerated with appropriate acid and alkali treatment. It should be remembered that exhausted resins release the trapped ions into the treated water.

To overcome this problem and to avoid exposure of the staff to potentially toxic acid and alkali solutions, we recommend the use of an electric deionizer for those units aiming for ultrapure water or for HDF and for online generation of replacement fluid.

Deionizers are used as final polishers for water already treated by RO or to decrease the total dissolved solids and conductivity of the feed water to the first stage of an RO system. By decreasing the load on the RO, the life of the membranes is increased, while the resins of the deionizer can be regenerated multiple times with acid and alkali. The disadvantages of deionizers include bacterial growth (infrequent with regular regeneration) and the discharge of large volumes of concentrated acid and alkali to the drainage system.

Electric deionizers use electrodes to adsorb ions in the water passing through a deionizer tank, which are then released and discharged to drain by reversing the polarity. There is no resin susceptible to bacterial growth and no requirement of chemicals which are hazardous.

In the Dutch study by Penne et al., the use of electric deionizers was the only intervention associated with lower bacterial contamination, while ultraviolet (UV) light, heat versus ozone, high- versus low-level disinfection, and SS versus cross-linked polypropylene (PEX) did not show significant benefit.

We recommend that an assessment be made of the following before designing a water treatment unit:

  1. The quality and possible contaminants of the source water. These should be calculated at maximum contamination during the year
  2. The amount of water needed. Assuming that each HD machine would work three shifts of 4 h each every day, 480 L of water/day will be needed per machine.

We recommend that incoming water should be treated for the following (pretreatment):

  1. Sediment filtration to remove suspended impurities
  2. Activated carbon filtration to remove chloramine
  3. Softener or deionizers.

We suggest that UV light of 180–400 nm can be used as an alternative to activated carbon.

Activated carbon or charcoal filters act to remove the chlorine and chloramines in water, which exist as a consequence of disinfection by chlorination carried out by the civic authorities. The water without chlorine is conducive to growth of bacteria and formation of biofilm with liberation of endotoxin because of the porous structure and nutrient-rich environment of granular activated carbon (GAC) tanks. Additional problems encountered with the use of GAC filters are increased loss of pressure head, difficulty in ensuring adequate empty bed contact time (EBCT) regeneration costs, and unpredictable chlorine breakthrough. If a softener follows the carbon filter, the resin is an additional rich ground for growth of bacteria, in water from which the chlorine has been removed.

We recommend that two carbon tanks in series, each with an iodine number of >1000 and an EBCT of >5 min (total 10 min) should be used.

The replacement of carbon tanks by sodium metabisulfite dosing and short UV radiation reduces the chances of bacterial growth while producing an equally efficient removal of chlorine and chloramines. The UV dosage required for dechlorination depends on total chlorine level, ratio of free versus combined chlorine, background level of organics, and target reduction concentrations. The usual dose for removal of free chlorine is 15 to 30 times higher than the normal disinfection dose. Typically, a UV radiation of 600 mJ/cm2 at 254 nm UV light wavelength will lower the chlorine concentration by a factor of 10, but this may vary according to the concentration of other organic and inorganic contaminants. This will usually provide adequate reduction of chlorines and chloramines from 15 and 5 ppm, respectively.

An additional advantage of activated carbon is the removal of pesticides and herbicides which may contaminate the raw water. No data currently exist about the removal of these compounds by metabisulfite and UV light.

We recommend that a SS (Grade 316 or 316L) or medical-grade PVC (uPVC) water tank be used for water storage. The tank must have deaeration valve and drain facility at the bottom so that complete water could be drained out. It should have an airtight lid and a tapered bottom to ensure complete drainage of water or disinfectant.

We recommend that all pipelines, valve joints, and connectors after RO system should be SS (Grade 316 or 316L) or medical-grade PVC or PEX.

We recommend that all pipelines should be flexible without joints or sharply angled bends. And that, machine to pipeline connections should be of the push–pull type.

We recommend that bends and blind loops should be kept to a minimum and that the distribution system should be a continuously circulating loop.

We suggest that the internal material finish of SS pipes should have an arithmetical average surface roughness of not >0.8 μm, which may be achieved by mechanical and electro-polishing techniques.

We recommend that the velocity of water in the pipe should be >0.9 m/s. Higher flows are associated with greater shear stress and decreased biofilm formation. As flow meters display flow in liters per hour, the velocity should be calculated using the formula: Velocity = Flow/cross-sectional area.

We recommend that all pipelines, valve joints, and connectors after RO system should be SS (Grade 316 or 316L) or medical-grade PVC or PEX.

If PEX is used, it should not be exposed to direct sunlight. We recommend that bends and blind loops should be kept to a minimum and that the distribution system should be a continuously circulating loop.

We suggest online 0.22 μ membrane filter or an UF unit and UV light after RO.

The ultrafilters or 0.22-μ filter should be placed after the UV lamp and at the start of the loop.

UV light while killing bacteria does not remove bacteria or their breakdown products from the water and these can produce inflammation. Ultrafilters typically retain dead and viable bacteria, endotoxin, and most bacterial products.

UV light is also suggested after activated carbon filter and before RO (see above).

Note: PEX tubings and joints may be damaged by continuous exposure to UV light.

For monitoring, we recommend the following:

  Chemical Purity Top

We recommend that the water leaving the pretreatment system be tested daily at least for residual total chlorine and hardness. This testing can be carried out in a side laboratory or in the RO room using orthotoluidine reagent and a comparator disc for chlorine and using ethylenediaminetetraacetic acid -based comparator with Eriochrome black T indicator or colorimetric paper strips for hardness. If the water leaving the second GAC tank contains chlorine >0.1 ppm, we recommend that the activated carbon be changed and that dialysis be stopped during this time.

We recommend that the pre-RO water should have a hardness not exceeding 17 ppm.

Apart from producing hemolysis in patients, free chlorine and chloramines damage thin film composite RO membranes. Water with increased hardness can scale RO membranes, reducing the permeate flow and eventually the life of the membrane.

Procedure of testing for residual chlorine using Aqua Chlorine Comparator™:

Fill the test tube provided with water from the post carbon tank sampling port about three quarters full. Add 1 drop of orthotoluidine reagent and mix. Stand the tube in the comparator and rotate the disc till the color in the tube solution matches that of the disc window. Read off the value of total chlorine from the comparator adjacent to the matching window.

Procedure for testing hardness using Aquasol (Rakiro Biotech Systems Pvt Ltd, Thane Industrial Estate, Thane, Maharashtra India) hardness kit:

  • Take 10 ml of water to be tested in the test jar
  • Add one measure of the reagent TH1 (total hardness)
  • Mix contents well to dissolve. The solution develops a blue green color
  • Add 20–25 drops of the reagent TH2 and mix well. The color turns red
  • Now add drop wise the reagent TH5, mixing the solution after each drop, counting the number of drops until the color just changes from red to blue
  • Calculations: Total hardness in mg/L or ppm = 25 × (drops of TH5).

Both hardness kit and chlorine comparator are available from M/s Achala Engineering and Electronics, Thane Industrial Estate, Kalwa, Thane – 605.

Website –

We recommend the use of online conductivity meters after deionizers and RO. There should be visible and audible alarm for improper conductivity in the dialysis technician's station. The alarm should lead to stoppage of water beyond RO. The water should re-start only after adequate conductivity is achieved.

We recommend that water should be sampled from the following points of a system for microbiological testing:

a) Feed water

b) Post carbon filter

c) At the point where the water leaves the RO machine, before the holding tank (indirect system)

d) If an RO water-holding tank is present, where the water leaves the tank

e) At the end of the return line of the RO water distribution loop

f) Dialysate.

We suggest that the following additional points be sampled for water testing, especially in cases of breakthrough positive cultures as these are possible points of contamination:

a) Post softener

b) Where water enters into the dialyzer reprocessing system

c) Where water enters equipment used to prepare bicarbonate and acid concentrate

d) At random points where the dialysis machine is hooked up to the product water loop.

The treated water sample should be sent for detailed chemical analysis to an independent laboratory having adequate instrumentation for testing at least once in 3 months. The results should be a mandatory part of the record system.

  Microbiological Purity Top

This should be checked once every 30 days to achieve the standards, as shown in [Table 2].

We recommend that pour plate method on nutrient poor medium should be used. Incubation should be at room temperature (20°C–24°C) for 7 days. Samples for culture will have the highest yield just prior to a disinfection cycle.

  Endotoxin Levels Top

We recommend that endotoxin levels should be checked once in every 30 days to achieve the standard, as shown in [Table 2]. Samples for endotoxin will have the highest value just after a disinfection cycle.

We recommend that each component of the water treatment system must be thoroughly cleaned and disinfected as per the manufacturer's recommendation. After disinfection, it is essential that the sterilent is completely removed before the treated water is used for dialysis.

We recommend the following for cleaning of RO membranes:

  1. Membranes should be taken offline and the system should be shut down during the process. The flow of reject should gradually increase if the process is successful
  2. Membranes should be backwashed at low pressure using an external pump and flow in the same direction as normal operation. Each membrane should be cleaned individually

The following cleaning solutions are required:

  • Sodium tripolyphosphate and sodium edetate at a pH adjusted to >10 to remove calcium scales and low-level organic foulants
  • 2% citric acid (no pH adjustment required) to remove calcium carbonate, metal oxides, and inorganic colloidal compounds. This also provides disinfection
  • A water wash is recommended between the cleaning solutions
  • A final disinfection using 1% peracetic acid or 2% formalin (to be treated for at least 6 h) is recommended
  • We recommend discarding the first 200–250 L of water or permeate over 1 h after a cleaning and disinfection cycle. The pH of permeate should be confirmed before using.

[Table 3] contains a list of the cleaning agents required for specific membrane foulants. It should be noted that membrane cleaning is never complete because of the vast pore number and membrane surface area, and that preventive maintenance and properly designed pretreatment is therefore the key.
Table 3: Membrane fouling and cleaning

Click here to view

The following steps are recommended for cleaning and disinfection of the distribution system:

  1. The storage tank should be filled with 50–100 L of 1% sodium hypochlorite (only PVC is compatible with sodium hypochlorite) or peracetic acid at the recommended dilutions (see section on dialyzer reprocessing)
  2. After allowing a contact time of 30 min, the solution is circulated in the loop for 20 min and drained followed by a complete rinsing with water that is discarded until a negative test with a starch iodide paper or a conductivity equal to that of the feed water is obtained.

We suggest that peracetic acid and heat be the preferred methods of disinfection.

We recommend that if peracetic acid is used for disinfection, a concentration of >1000 ppm should be used (documented by a semi-quantitative reagent strip) and a circulation time of at least 30 min is used. The disinfectant should be flushed repeatedly with 50–100 L of water until the effluent contains <2 ppm of hydrogen peroxide (documented by a semi-quantitative reagent strip).

Water heated to 85°C or greater is ideal for disinfecting systems of SS or PEX. It is important that the water circulating in the loop maintains this temperature, hence an automatic control system for the heaters is required. As no chemical is used, this method can be used daily and has been shown to completely prevent the formation of biofilm in a system of 316 SS disinfected daily over a period of 2 years. [Table 4] shows the incompatibility of disinfection methods with the material, of which the tank and pipeline are constructed. The use of physical or chemically incompatible disinfection methods may result in damage to various components of a distribution system and provide areas for bacterial colonization and biofilm formation.
Table 4: Disinfection agents and compatibility

Click here to view

We recommend a log for documenting the performance of the water treatment system components, indicating the maintenance done on each component.

A suggested schema is shown in [Table 5]. [Figure 1] shows the planning of a water treatment plant for a stand-alone unit.
Table 5: Maintenance of water treatment system

Click here to view
Figure 1: Schematic diagram of water treatment

Click here to view

Limulus amoebocyte lysate assay for endotoxins

  1. Add 0.1 ml of reconstituted (as instructed in package insert) Pyrotell® (C. No. G2003/G2125) (0.03/0.125 EU/mL)
  2. Add 0.1-ml test specimen or control (C. No. E0005 – positive control or C. No. W0504 LRW – limulus amoebocyte lysate (LAL) reagent water – negative control)
  3. Mix vigorously (vortex) for 20–30 s
  4. Place the reaction tubes at 37 ± 1°C water bath for 60 min
  5. Remove reaction tubes and invert the tubes in one smooth motion
  6. A positive test is indicated by the formation of a gel which does not collapse when the tube is inverted by 180°.


  1. Nexus, Division of Span Diagnostics, 173-B New Industrial Estate, Udna, Surat – 394 210. Cell: 9825831746, Fax: +91 261 28679319
  2. Lonza India Private Limited, 2nd Floor, Krishnama House, 8-2-418, Road No. 7, Banjara Hills, Hyderabad – 500034, Tel: +91 40 4123 4000, Fax: +91 40 4123 4090, Cell:+91 87 9099 6211.

Microbial culture

  1. One hundred milliliter of the treated water is collected in sterile and pyrogen-free container*

  2. For endotoxin testing, a sterile pyrogen-free disposable hypodermic syringe can be used as a sterile pyrogen-free container and the sample is transported to the lab immediately

  3. Pour 1 ml sample of water on Soya triptone agar (nutrient deficient), McConkey agar, and nutrient agar plates.

  4. Suggested source: Himedia Labs (

    • Place in an incubator at 37°C and 25°C, and check for the bacterial colony formation after 48 h and 7 days
    • Colonies are counted in the plates with positive growth and expressed as the CFU per ml.

Test for microelements in reverse osmosis water

  1. One hundred milliliter of the treated water is collected in a glass-stoppered bottle (Borosil) previously washed in the following way for proper decontamination from microelements
  2. Bottle is washed twice with mild detergent solution followed by repeated washing with treated/pyrogen-free water
  3. The washed bottle is oven dried
  4. The bottle with hydrochloric acid (HCl) is diluted in treated/pyrogen-free water and then keep overnight under 0.1 N HCl
  5. Repeatedly wash with treated/pyrogen-free water
  6. Wash the bottle with the sample RO water ten times with vigorous shaking
  7. Drain out the washing water
  8. Collect the sample of treated water (100 ml) in bottle for analysis. Close the stopper immediately
  9. Estimation is done for the following analytes: aluminum (Al), arsenic (As), barium (Ba), beryllium (Be), bismuth (Bi), calcium (Ca), cadmium (Cd), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), potassium (K), lithium (Li), magnesium (Mg), manganese (Mn), nickel (Ni), phosphorus (P), lead (Pb), sodium (Na), palladium (Pd), rubidium (Rb), antimony (Sb), tin (Sn), titanium (Ti), vanadium (V), zinc (Z), chloride, thallium (Tl), nitrate, fluoride, and sulfate.


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5]


Print this article  Email this article


Indian Journal of Nephrology
Published by Wolters Kluwer - Medknow
Online since 20th Sept '07