Sreelakshmi and Eldridge have raised concerns in their case report regarding the safety of cell salvage and autologous blood transfusion through a leucocyte depletion filter in obstetric practice.  In their case, invasive blood pressure monitoring was instituted during the procedure, rather than at the start.  We feel that in cases where large volume blood loss is anticipated, an arterial line is invaluable for the rapid detection of haemodynamic instability and frequent blood sampling.

During the case reported by Sreelakshmi and Eldridge the amniotic fluid was aspirated from the surgical field and then discarded. A prospective randomised study compared the efficiency of cell salvage at removing amniotic fluid in two groups. In the first group two suction devices were used and this was compared to the use of one suction device.  There was no statistically significant difference between the two groups, which would suggest that the use of only one suction device is safe and should in theory increase the efficiency of red blood cell recovery [1].

We agree that bacterial contamination and amniotic fluid embolus are unlikely causes for the hypotension in these cases. Various studies have investigated the autologous transfusion of microbiologically contaminated salvaged blood and demonstrated no adverse outcomes or increase in postoperative infectious complications [2-5].  To date, there have been no proven cases in the literature of amniotic fluid embolus caused by re-infusion of salvaged blood [6].  

When considering the cause of hypotension, one of the relative contraindications to pressurising salvaged blood is the risk of air embolus due to the presence of air in the bag of autologous blood.  An air embolus may have accounted for the cardiovascular instability demonstrated by the patient in the case report. Another important consideration is the administration of allogeneic blood products during the management of massive obstetric haemorrhage and coagulopathy.  Immediate or delayed adverse reactions to these blood products are more common than reactions to allogeneic or autologous red blood cells and again could have accounted for the hypotension.  Were there any other signs of a blood product transfusion reaction or anaphylaxis such as pyrexia, bronchospasm, urticaria or rash? It would appear that there was a temporal relationship between the administration of cell salvaged blood and the episodes of hypotension in both cases which is most likely to be attributable to bradykinin.  

Re-transfusion of cell salvaged blood through a LDF is well established and has a long history in surgery for urological malignancy [7]. Cell salvage has shown to be safe in a 5-year retrospective review of adverse events associated with blood transfusion.  They examined over 27 000 cases where salvaged, pre-donated autologous or allogeneic blood was transfused, and found the incidence of adverse events with autotransfusion to be 0.027% compared to 0.14% with allogeneic blood transfusion [8].  The Cochrane Collaboration published a meta-analysis in 2006, which found that cell salvage is safe and efficacious at reducing the need for allogeneic blood transfusion [9]. Both case reports have emphasised that there are many causes of hypotension associated with massive obstetric haemorrhage, and the transfusion of autologous and allogeneic blood products.  In most cases the cause of hypotension will be multifactorial, but when transfusing salvaged blood through a LDF, bradykinin is an important differential diagnosis. These reports of hypotension associated with cell salvage and a LDF have been important to highlight the potential problem of bradykinin but should not deter our enthusiasm for the use of cell salvage as an important method of blood conservation.

A. Ashworth 

N. O’Keeffe

Manchester Royal Infirmary, Manchester, UK.

A. R.Jones 

Royal Oldham Infirmary, Oldham, UK

References

1.    Sullivan I, Faulds J, Ralph C.  Contamination of salvaged maternal blood by amniotic fluid and fetal red cells during elective caesarean section.  British Journal of Anaesthesia 2008; 101:225-9

2.    Feltracco P, Michieletto E, Barbieri S, et al.  Microbiologic contamination of intraoperative blood salvaged during liver transplantation.  Transplantation Procedures 2007; 39: 1889-91

3.    Bowley DM, Barker P, Boffard KD.  Intraoperative blood salvage in penetrating abdominal trauma: a randomized, controlled trial.  World Journal of Surgery 2006; 30:1074-80.

4.    Ozmen V, McSwain NE Jr, Nichols RL, Smith J, FlintLM.  Autotransfusion of potentially culture-positive blood (CPB) in abdominal trauma: preliminary data from a prospective study.  Journal of Trauma 1992; 32: 36-9.

5.    Boudreaux JP, Bornside GH, Cohn I Jr.  Emergency autotransfusion: partial cleansing of bacteria-laden blood by cell washing.  Journal of Trauma 1983; 23: 31-5.

6.    The Association of Anaesthetists of Great Britain and Ireland (AAGBI) Safety Guideline – Blood transfuion and the anaesthetist: Intra-operative cell salvage.  http://aagbi.org/publications/guidelines/docs/cell%20_salvage_2009_amended.pdf.2009 (accessed 27/08/09)

7.     Nieder AM, Carmack AJK, Sved PD, Kim SS, Manoharan M, Soloway MS.  Intra-operative cell salvage during radical prostatectomy is not associated with greater biochemical recurrencerate.  Urology 2004; 65:730-734.

8.    Domen RE. Adverse reactions associated with autologous blood transfusion:evaluation and incidence at a large academic hospital.  Transfusion 1998; 38: 296-300.

9.     Carless PA, Henry DA, Moxey AJ, O’Connell D, Brown T, Fergusson DA.  Cell salvage for minimizing peri-operative allogeneic blood transfusion.  Cochrane Database Systematic Reviews 2006; 4: CD001888.

We thank Ashworthand colleagues for their comments.

We agreewholeheartedly that cell salvage is an excellent technique in the instance of obstetric haemorrhage, as well as in other circumstances. Our case report was not intended todeter individuals from using cell salvage, only to alert them that hypotensionmay, on rare occasions, be associated with the use of leucocyte depletionfilters. This is such a well-recognised problem in haematologicalpractice that bedside use of leucocyte depletion filters is no longer recommended [1, 2]. However we are not suggesting that eucocyte depletion filters should not beused, only that if hypotension does occur, it may be sensible to remove the leucocyte depletion filter.

It is true thatwe are not able to stare with certainty that the cause of the hypotension was the eucocyte depletion filter in this case, but we believe that this explanation is the best fit for theobserved events.

We are able to statewith a high degree of confidence that an air embolus was very unlikely to havebeen the cause; firstly, as although the bag and filter were pressurisedduring our first attempt to prime the giving set, at this stage the line wasnot connected to the patient and the line became blocked. On our second attempt, with a new giving set andfilter, the line was allowed to prime using gravity alone. At no stage wasautologous blood given to the patient through a pressurised line. Secondly, thetiming of the hypotensive events do not seem to be consistent with an airembolism. The blood pressure recovered rapidly when the infusion was stopped, and the hypotension reoccurred when the drip was recommenced. The line wasunder observation and no air was spotted.  Thirdly, no large changes were noted in theend tidal carbon dioxide during this period. Therefore, we feel air embolism was highlyunlikely to have been the cause of the hypotension.  

We agree thatbecause of the temporal relationship to the infusion of the cell salvagedblood, it is very unlikely that the hypotension was due to allogenic blood thathad been transfused earlier. We can also confirm that bronchospasm, urticariaand pyrexia were not noted at the time.

Although wethink it is not relevant to the current discussion, we do still discard fluid heavily contaminated with amniotic fluid before collection. Weconsider further evidence supporting safety is required as the benefits of the marginallyincreased volume of red cells collected during this early phase are likely tobe minimal. In addition, the fluid collected during this early phase is mostlikely to be contaminated by fetal red cells, especially if the placenta hasbeen transected in order to deliver the fetus. However, we would not have an issue with individuals who do choose tocollect all the fluid during surgery.

We also thinkthat the timing of the insertion of the arterial line was again not strictlyrelevant to the discussion. We had sufficient number of staff available thatthe arterial line could be inserted rapidly when required.

During the1990s, there were at least 390 reported cases of cell salvage without the useof leucocyte depletion filters in obstetric practice [3-5]. There were no reports of adverse consequences.However, during the last ten years, with data supporting the use of cellsalvage with leucocyte depletion filters, cell salvage in obstetric practice has become much morepopular. Therefore, we continue to use and commend the use of cell salvage and wecontinue to use leucocyte depletion filters. However, in our practice, if we were to see unexplainedhypotension in association with the use of a leucocyte depletion filter again, we would remove the leucocyte depletion filter from the circuit.

References

1. U.S. Food and DrugAdministration. Hypotension and bedside leukocyte reduction filters. 1999. http://www.fda.gov/cdrh/safety/hypoblrf.html

2. Shiba M, Tadokoro K,Sawanobori M, et al. Activation of the contact system by filtration of plateletconcentrates with a negatively charged white cell removal filter andmeasurement of venous blood bradykinin level in patients who received filteredplatelets. Transfusion 1997; 37: 457-462.

3. Jackson SH, Lonser RE. Safety andeffectiveness of intracaesarean blood salvage. Transfusion 1993;33: 181.

4. Rainaldi MP, Tazzari PL,Scagliarini G, Borghi B, Conte R. Blood salvage during caesarean section. British Journalof Anaesthesia 1998; 80:195-8.

5. Grimes DA. A simplified devicefor intraoperative autotransfusion. Obstetrics and Gynecology 1998; 72: 947-50.