Investigation of possible causes of massive hemolysis occurring at Toronto Western Hospital.
Subject: Hemodialysis (Complications and side effects)
Hemolysis and hemolysins (Health aspects)
Hemolysis and hemolysins (Research)
Blood circulation disorders (Causes of)
Blood circulation disorders (Research)
Authors: Tantalo, Franca
Holtzer, Carol
Pub Date: 07/01/2008
Publication: Name: CANNT Journal Publisher: Canadian Association of Nephrology Nurses & Technologists Audience: Trade Format: Magazine/Journal Subject: Health care industry Copyright: COPYRIGHT 2008 Canadian Association of Nephrology Nurses & Technologists ISSN: 1498-5136
Issue: Date: July-Sept, 2008 Source Volume: 18 Source Issue: 3
Topic: Event Code: 310 Science & research
Geographic: Geographic Scope: Canada Geographic Code: 1CANA Canada
Accession Number: 187615902
Full Text: This paper was presented at the Ninth Annual Symposium of CANNT held in Toronto in November 1984. It was first published in the CANNT Dialtec, (1985, February), 8(2), 6-8.

We have previously described four cases of massive intravascular hemolysis occurring in the hemodialysis unit at Toronto Western Hospital between December 1982 and April 1983.

Usually, when massive hemolysis occurs in a patient on dialysis, the cause is evident immediately as a gross error or failure in the system. The unusual feature about this epidemic is that the cause was not obvious and continued to defy investigation for a long time. The measures taken to try to discover the cause or causes are as follows:

1. Dialyzer re-use

A reason for ruling out the re-use process as a cause of hemolysis is that three of the patients were on new dialyzers while two were on re-used ones. Also, the dialyzers were from different companies. It seems, therefore, that the episodes of hemolysis had nothing to do with the dialyzer itself.

2. Incorrect composition of dialysis fluid

Severely hypotonic dialysis fluid can cause hemolysis, but two facts are adequate to exclude this as a cause in our cases. All the affected patients were receiving dialysate from the central delivery system. A gross error in composition would have affected all patients simultaneously. The dialysate was analyzed in three of the cases and was found to be correct.

3. Contamination of the dialysate with copper, chloramines, or nitrates

Analysis of the dialysate showed no traces of copper, chloramines, or nitrates. Blood tests performed at the time of hemolysis also showed no signs of these substances. The fact that all other patients on the same dialysis sessions receiving the dialysate were unaffected is further evidence against these causes.

4. Possible contamination of the dialysate entering the single patient stations

The function of this single patient station is to conduct the dialysate through the dialyzer and down the drain. The dialysate is not heated by the station, nor does the station add anything to the dialyzing fluid. It simply controls the flow rate and the pressure of the dialysis fluid. The machines were examined for the possibility that some copper part had been erroneously added to the machine that could have introduced copper into the dialysate and, hence, caused hemolysis. No such defect was discovered and, furthermore, analysis of dialysate samples and blood samples failed to reveal any traces of copper in the system.

5. Overheated dialysate

Overheating would have affected all patients simultaneously, not just one at a time. Dialysate temperature was checked and found to be correct.

6. Malocclusion of blood pumps

The five patients in the epidemic were being treated with four different machines at four different geographical locations within the dialysis unit. The blood pumps would be expected to damage the red cells of other patients dialyzed on the same machine. The pumps on all the machines were checked and no sign of malocclusion was found.

7. Damage of red cells from incorrect positioning of arterial fistula needles

Hemolysis caused by trauma to red cells generated by high negative pressure of malpositioned fistula needles has been described. Mechanical trauma of this kind would not be expected to induce methemaglobinemia and could, therefore, not explain two of our cases. All pre-pump arterial monitoring has always been in use in our unit ensuring that poor flow arterial needles are adjusted.

8. Possible intoxication of the extracorporeal circuit by the heparin or saline infusions

It was felt necessary to exclude the heparin and saline infusions as causes. Accordingly, in the last two cases, the whole extracorporeal circuit was preserved, including the remnants of the heparin and saline infusions. These were incubated against fresh red cells in test tubes in body temperature for a period of time. No hemolysis occurred. It seemed clear that neither one of these could be the cause.

9. Possible contamination of the dialysate by hypochlorite used for sanitation

It was pointed out to us that hypochlorite passing through the rest of the system could diffuse in small quantities into the open ends of the cul-de-sacs or "dead legs", which had been closed off distally. When rinsing was done with water, it would not necessarily remove all traces of hypochlorite that may have entered these dead legs if they remained shut off. Traces of hypochlorite might emerge from the dead legs into the dialysate pathway flowing to the patient station. In order to test this hypothesis, hypochlorite was mixed with dialysate to determine the minimum toxic concentrations of hypochlorite required to hemolyse red cells both for uremic and non-uremic patients. It turned out that the dead legs may contain a concentration of hypochlorite between 2 to 40 parts per million. When dialysate was run past these areas, it was not possible to detect hypochlorite in concentrations as low as one part per million. A further argument against the hypochlorite theory was the fact that two patients were being dialyzed on stations in the unit with a dialysate supply, which is not juxtaposed to any cul-de-sac.

10. Contaminated blood tubing sets

One factor that was in common with the first four cases of hemolysis was the use of Extracorporeal brand blood tubing sets. The manner in which the hemolysis occurred in these patients suggests that a hemolytic substance with oxidative properties was delivered into the extracorporeal system gradually during the course of dialysis. This suggests a gradual infusion or leaching process of a hemolytic factor into the extracorporeal circuit.

Summarizing the results of Dr. Pierre Blais' investigations, it appears that the extracorporeal bloodlines were contaminated with a wide variety of extraneous impurities. Visible particulate matter was seen in the bloodlines and identified as chips of wood, pieces of blue paint, black particulates and a number of unidentified chemicals. Some of the bloodlines also contained undesirable amounts of plastic adhesive, cyclo-hexanone, which is used to join the plastic wherever joins are needed. Although we have not yet identified the actual substances that may have caused the hemolysis, it seemed very likely that hemolysis was caused by sporadic contamination of the bloodlines in these four cases. If these bloodlines contain a variety of miscellaneous substances that should not be present, then it is entirely possible that one in 1,000 sets of lines may be more heavily contaminated by some substance, as yet not identified.

The hemolytic case five, during dialysis using Cordis Dow bloodlines naturally throws doubt on this theory. Could it be that the bloodlines used in all the hemolytic episodes were derived from the same extruder, or is it possible that the adhesive used to join them all came from the same source? As yet, we regard the mystery still as unsolved. Plans are under discussion to further investigate unused Extracorporeal blood tubing sets from Toronto Western and from a hospital in Syracuse, New York.

Following the occurrences of hemolysis at Toronto Western, an inquest was held to determine the cause of death of the second case. The purpose of the coroner's inquest was not to find fault or to lay legal responsibility on any one party, but to find circumstances related to the death and make recommendations to prevent further episodes. We cannot possibly try to describe the details of the inquest, including all the evidence and discussions had in such a short period of time. Nor can we impart the effects of an inquest on the unit itself or even more so the individual nurse's reactions and feelings to being on the "stand'.

The jury's verdict and recommendations are the following:

1. We recommend that the Health Protection branch of Health and Welfare Canada continues investigation of PVC blood tubings as manufactured by the Extracorporeal Company, which is a division or subsidiary of Johnson and Johnson, in relation to hemolysis complications with respect to the incidences at the Toronto Western Hospital from September 1982 until April 1983. Funding for this project to be provided by the Government of Canada.

2. To avoid grossly inferior medical devices that have been imported by Critikon Canada Inc., we recommend that the Ministry of Health and Welfare set standards for plastic tubing and other plastic medical devices; such standards to be mandatory upon both domestic and imported product. Failure of any manufacturer or importer to meet said standards would be subject to punitive measures and/or penalties as established by the government authorities.

3. We recommend that a permanent government inspector be appointed with the responsibility to inspect both domestically and imported goods, thus ensuring the products meet the specifications that will be established, as in recommendation 2.

4. We recommend that hemodialysis departments inform blood laboratory departments of the vital importance to them of knowing if a patient's blood is hemolysed. Improved communication between the labs involved and the ward--as within four hours three tests indicated "grossly hemolysed" and "marked hemolysis" were used.

5. We recommend that when a death occurs in a hemodialysis unit where the cause of death cannot be positively established, an autopsy should be performed. If consent is not given by the family, a coroner should be informed immediately to secure his approval or direction for autopsy.

6. We recommend that hospital personnel involved with hemodialysis units should be instructed on visual recognition of manufacturing defects and/or particulate in disposable plastic medical devices.

7. We recommend the formation of a consumer/user/professional society or association be established by those involved with hemodialysis. Purpose: to protect buyer and user, plus lobby government bodies, as well as publish and print pamphlets, brochures and other information for new dialysis patients and their families.

The quality of blood tubings, as seen during this inquest, is totally unacceptable, as well as being appalling.

A handsome and generous donation from Johnson and Johnson towards the formation and charter of such an organization/society would indicate some measure of their good will.

8. We recommend that dialysis patients should have intensive examinations and assessment of blood and general condition more frequently than in the past.

9. We recommend that only new dialyzers be used in hemodialysis units.

10. We recommend that any and all cases of hemolysis diagnosed in a hemodialysis unit in Canada be reported immediately to Health and Welfare Canada, Health Protection branch.

11. We recommend that for each dialysis, a responsible person record brand and lot number of the tubing used in a patient's chart.

12. When the Ministry of Health and Welfare issues an "alert" on any product, we recommend that legislation be made prohibiting any person.

13. We recommend that any hospital in Canada with a dialysis unit with a dialysate delivery system with any "dead legs" should be flushed out during the regular sanitation process, or otherwise eliminate them.

14. We recommend guidelines on the storage of bloodlines. The temperature of the storage area/room should not exceed 30 to 40 degrees C. Nor should bloodlines be stored adjacent to odour-emanating materials. Cartons should clearly indicate storage warning and information.

15. We recommend that all records pertaining to any patient be retained for a full 12 months after the date of the last entry.

As a result of the recommendations and the exercise of the investigations and questions by numerous officials, we have initiated the following changes in nursing protocols and procedures:

1. Thorough recording of the supplier and lot number of the dialyzer and the bloodlines.

2. All the disposable packaging used for each dialysis is kept in a bag at the bedside until the dialysis is discontinued.

3. All treatments, checkouts, or medication given must be signed off by the nurse performing the task.

4. No advance preparation of the supplies is done. All disposables to be used are unpackaged at the bedside just before a patient is to be started on dialysis.

Along with these changes, we have learned of other practices of which to be aware. Packaging of blood tubing should be observed. The punch-through or window-type is the most primitive form of gas-compatible packaging for it allows the ingress of dirt or other contaminants. This type was used by Extracorporeal. The permeable punch-out covered by a porous patch is the state-of-the-art in packaging. This type allows no ingress of fluid or dirt. Cobe packaging is an example of this. The double-sided package incorporates plastic on one side and porous paper on the other. This type is frequently used and is a good gas-compatible package. Hospal and Drake Willock are good examples. Very few packages allow a noticeable reseal of the bag. The haessen seal was seen in the Extracorporeal tubing, as the packages were different lengths and widths.

Storing of disposables is quite critical. Bloodlines are like sponges and absorb chemicals or odours surrounding them. They should be stored away from industrial activity such as loading docks or trucks. Some samples of blood tubings were found to have high levels of diesel, as they were stored in a poorly vented loading area. With all of the above information, all staff at the Western are now instructed to visibly inspect all disposables. Any obvious defects and odour should be reported and the product not used.

Nurses are educated about hemolysis including the spinning of blood if it is suspected. We stressed that the lessons learned in the earlier days of hemodialysis cannot be forgotten or put aside. All nurses must be made aware of all potential dialysis complications and their treatments even if the complication is a rare one.

About the author

Franca Tantalo worked in dialysis for more than 30 years and was a member of CANNT for most of her career. She worked on the Standards of Practice for Nephrology Nursing and has published many articles. She also worked for Fresenius Medical Care as a Clinical Applications Specialist and, in her honour, Fresenius supported a bursary for graduate study in nephrology in her name. Franca passed away in August 2004 after a long battle with cancer (CANNT Journal, 14(3), 52.)

By Franca Tantalo, Head Nurse, and Carol Holtzer, Clinical Teacher
Gale Copyright: Copyright 2008 Gale, Cengage Learning. All rights reserved.