279

THE HAZARDS OF BLOOD TRANSFUSIONBy H. F. BREWER, M.D., B.Ch.(Cantab.)

Clinical Pathologist to St. Bartholomew's Hospital;Medical Officer to the Greater London Red Cross Blood Transfusion Service

Blood transfusion used correctly is a life savingmeasure and its reputation and use have steadilyincreased since the I939-45 war. Its growth andthe present important role it plays in the practiceof medicine have necessitated the State under-taking full responsibility for the provision of bloodand blood products and of standard apparatus withwhich to collect and administer them. Last yearthe National Blood Transfusion Service distributedover 700,000 bottles of blood and in the Londonarea the specialised Red Cross Service providedover 4000 donors purely for fresh blood trans-fusions. The spectacle of several patients in aWard having simultaneous transfusions is afamiliar one in the round of one's hospital duties.Nevertheless, in part owing to its ready availability,it is probable that a great deal of blood is givenunnecessarily, and because blood can prove alethal fluid and its administration carries certainrisks, this aspect is particularly relevant in anydiscussion of the associated hazards. We have allseen blood given when the haemoglobin levelhas not really indicated it and where no accounthas been taken of the patient's own potentialhaemopoietic regenerative powers, especiallywhen aided by generous diet, iron, folic acid orliver extract. Again, whole blood is sometimestransfused when plasma or serum or a plasmasubstitute (dextran) would achieve the same pur-pose or even be preferable. The indiscriminateuse of blood by subjecting patients unnecessarilyto some of the specific hazards is thus of itself asignificant danger in blood transfusion and onethat can only be eliminated by careful selection ofthe recipient and of the fluid to be administered.In the larger hospitals this is probably best assuredby the appointment of a Transfusion Officer.The hazards of blood transfusion may be listed

as follows--(I) Haemolytic reactions dependent on intra-

vascular haemolysis or on the transfusionof blood already lysed (extravascularhaemolysis).

(2) Sensitization of the recipient.(3) Use of the infected blood.

(4) Circulatory overloading.(5) Air embolism.(6) Transmission of disease from donor to

recipient.(7) Febrile and allergic reactions.(8) Miscellaneous-e.g. Transfusion throm-

bophlebitis. Haemosiderosis.This is a somewhat disturbing list and fortunate

is the man who has had a long association withthis form of treatment and has not seen it cause afatality or harm a patient whom it was hoped tobenefit.

i. Haemolytic reactionsAs already indicated these may be due to the

intravascular haemolysis of blood or to the trans-fusion of blood already lysed by some extravascularcause.

Intravascular haemolysis is the sequel of anincompatible transfusion and in most cases isdependent on the destruction of the donor's redcells by the action of specific iso-antibodies inthe recipient's circulation. The destruction maybe immediate or delayed, the severity of thehaemolytic reaction running parallel with therapidity of haemolysis. The significant anti-bodies concerned are anti-A and anti-B of theLandsteiner system and anti-D of the Rhesussystem. Between them these are responsible forpractically all the haemolytic reactions dependenton intravascular haemolysis of the donor cells.Anti-Kell is probably the only anti-body of theother group systems which has to be seriouslyconsidered.

Generally speaking an ABO incompatibilityproduces a reaction that is more acute and violentthan one due to the Rhesus factor. With the latterthe symptoms are usually (but not invariably)more silent and the appearance of jaundice afterabout I2 hours may be the first indication. Aslow drip rate of administration is known to renderthe symptoms of incompatibility less obvious.They are also masked under general anaesthesia.Here, because no typical picture of incompatibletransfusion is revealed, vigilance is particularly

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

280 POSTGRADUATE MEDICAL JOURNAL June I956important. It is good practice to record the pulserate and blood pressure in an anaesthetized patientat the commencement of a transfusion and onchanging to a fresh bottle of blood, and then at 5minute intervals for 15 minutes. If there is asharp rise in pulse rate and/or a rapid fall in bloodpressure, with perhaps some cyanosis, unexplainedby any event or manipulation in the operationfield, the transfusion should be terminated.Of all the protean and well-known symptoms

produced by the transfusion of incompatible bloodsevere lumbar pain is especially significant since itmay be one of the earliest warnings of haemolyticreaction. If it occurs the transfusion should bediscontinued until the possibility has been excludedby repetition of the grouping tests.The development of reliable and sensitive

techniques both for blood grouping and directmatching and the ready availability of potentgrouping sera since the War have greatly reducedthe incidence of incompatible transfusions due toimperfect serological methods.The direct matching test between donor's red

cells and the serum of the recipient is the finaland essential precaution against incompatibility,and provided a satisfactory technique is employedit can be relied upon to detect not only ABO andRh incompatibility but also any of the rarerantigen-antibody reactions, although their identitymay not be known at the time. Performance ofthe direct matching test, which should never beomitted, carries a heavy responsibility, since it isalmost the only laboratory investigation in whicha mistake may prove fatal. In the writer's opinionit should for safety only be carried out by someonewith laboratory training. For routine purposes atwo tube technique in which donor cells in salineand in 20 per cent. bovine albumin are put upagainst the recipient's serum is very reliable andpreferable to a tile method. The procedure is asfollows-a few drops of donor blood are removedfrom the transfusion bottle (with sterile pre-cautions) or from the pilot bottle and the cellswashed twice with saline. A drop of 2-5 per cent.suspension in saline of these washed donor cells isplaced in each of a pair of precipitin tubes with aPasteur pipette. To both is added a similar sizeddrop of the patient's serum and mixing effected bytapping the tubes with the finger. One tube isplaced in the incubator at 37°C and the otherallowed to stand at room temperature. At theend of one and a half hours the supernatant salinein the incubated tube is cautiously pipetted off,discarded and replaced by a drop of 20 per cent.bovine albumin, without disturbing the under-lying red cell sediment. The tube is put back inthe incubator for a further half an hour. After

this period the red cell deposit is removed fromboth tubes with a Pasteur pipette, gently spread ona slide, and examined for agglutinates under thelow power of the microscope. In urgent circum-stances a drop of a 2-5 per cent. suspension ofwashed donor cells in saline is placed in oneprecipitin tube and a drop of a similar strengthsuspension in 20 per cent. bovine albumin inanother. To each is added a drop of the patient'sserum. After tapping to mix, the tubes areplaced in the 37°C incubator (or preferably awater bath) for at least 15 minutes. The tubesare then centrifuged just fast enough (Iooo r.p.m.for one minute) to bring down the cells to thebottom of the tube. The sediment is then trans-ferred to a slide and examined as described.Suspension of the donor cells in a protein mediumis important in order to detect the presence of theso-called incomplete antibodies (mainly anti-Dof the Rh system) which are a relatively commoncause of potential incompatibility and whichwould be missed if a saline suspension alone wereemployed. The latter detects more effectivelythe antibodies anti-M, anti-N, anti-P anti-Lewisand non-specific cold agglutinins. For certaincases who are liable to have an immune isoanti-body in their serum (Rh antibody is the mostsignificant)-sometimes referred to as ' dangerousrecipients,' e.g.-Rh-negative patients, patientswho have had previous transfusions or transfusionreactions, mothers of children where there isreason to suspect haemolytic disease, cases ofacquired haemolytic anaemia-it is wise toaugment the two tube test by the Indirect Coombs'test. This is more sensitive than the albumintechnique in detecting traces of immune antibodiesbut it cannot alone be relied upon for demon-strating incompatibility dependent on naturally-occurring anti-A and anti-B agglutinins.

Haemolytic reactions dependent on the destruc-tion of the patient's own red cells caused by thetransfusion of incompatible plasma, e.g. the use ofa Group O donor for recipients other than Group0, do occur. In most cases the concentration ofthe anti-A and anti-B agglutinins in the plasma ofthe Group O donor blood is diluted to a safe levelin the recipient's circulation, but some Group Odonors possess a high titre of these agglutinins,and particularly with massive transfusions given toa small and exsanguinated patient other thanGroup 0, can cause some intravascular haemolysisof the recipient's red cells. This seldom resultsin a frank haemolytic reaction comparable withdestruction of incompatible donor cells and if thepeace-time practice of only using homologousgroup ABO blood for a recipient is adhered to, ahaemolytic reaction due to incompatibility of theplasma of the donor blood is extremely rare. It

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

282 POSTGRADUATE MEDICAL JOURNAL June I956does not demand a direct matching test betweendonor's serum and recipient's cells. The responsi-bility of ear-marking Group O donors with hightitre agglutinins should lie with the TransfusionService. Bearing on the question of incompatibleplasma is the production of anti-A and anti-Bimmune bodies in Group O individuals followingthe injection of horse serum and prophylacticbacterial vaccines. Such immune antibodies arecapable of reacting powerfully with A or B cells.Safety lies in ensuring that a Group O donor whohas had for example anti-tetanus serum or pro-phylactic TAB vaccine in recent months is usedonly for Group O recipients and not as a ' uni-versal' donor.

Extravascular haemolysis. Haemolytic reactionsdue to massive extravascular destruction of thedonor red cells may result from the use of time-expired stored blood, and of blood that has beenaccidentally overheated or frozen.

Haemolytic reactions whether due to intra-vascular or to extravascular haemolysis are usuallythe result of gross blunders. Blood of an in-compatible group is given because of a clericalerror, or confusion between patients with similarnames. Haste in blood grouping and the directmatching test-or omission of the test-are muchmore significant causes than the action of obscureantibodies. In most hospitals time and moneyare better spent on improving transfusion organiza-tion rather than on the development of elaboratecompatibility techniques-provided that homo-logous ABO and Rh blood is selected, and areliable direct matching test, such as that alreadydescribed, is performed. An additional label putover the cap of a bottle of direct matched bloodand bearing the caption' Stop ! Have you checkedfull name of patient and Ward' on the compati-bility label is a wise injunction. If blood has topass through several hands before being issued forand given to a particular patient its identity shouldbe checked en route in the same way as a dangerousdrug. Even in urgent cases, the initial use of aplasma substitute will practically always tide thepatient over until grouping tests can be completed.Collapse is due more to lack of blood volume thanlack of red cells.

Nursing staff should be told of the danger ofoverheating blood; they seldom realize that bloodis damaged by exposure to temperatures above38°C. It is a wise rule for anyone carrying outtransfusions not to allow a bottle of blood to beimmersed in warm water unless the temperatureof the water has been first assessed by his ownhand. In practice, with certain exceptions-e.g.a high titre of cold agglutinins-preliminarywarming of the blood when given at a slow driprate is unnecessary.

Blood which has thawed after being frozen isequally dangerous. Such blood usually showsobvious haemolysis to the naked eye. Accidentalfreezing is avoided by fixing the thermostat control-ling the temperature of the storage refrigerator to arange between 2°-4°C and by charting the dailyrecords of a maximum and minimum thermometerplaced in it. Generally speaking domestic re-frigerators are not satisfactory for storing bloodsince temperatures vary from one part of thecabinet to another.

Before concluding the subject of haemolytictransfusion reactions there is now some comfort tobe derived from the fact that even in severeexamples, when anuria has developed, the regimeadvocated by Bull and his colleagues has proved tobe a great advance in reducing the mortality,previously associated with these disasters. Essenti-ally this treatment (which aims at resting thekidney as much as possible during the stage ofimpaired function) consists of a protein-free,mineral-free diet containing an adequate numberof calories and having a water content of onlyIooo c.c. per day (corresponding to the amountlost via the lungs, skin and faeces). It is admin-istered, owing to its unpalatability (an emulsion ofglucose and arachis oil) through a permanentlyindwelling stomach tube at a drip rate and con-tinued until the patient commences to pass ioooc.c. of urine a day. This treatment which shouldbe started at once is the only one which has so farimproved the prognosis in the anuric patient;provided the kidneys were previously healthy ityields a high incidence of recoveries. If it cannotbe instituted where the patient has had thetransfusion, he should be transferred to a hospitalwith experience of it.

Administration of alkalis and attempts to flushthe kidney by pushing fluids at this stage areharmful and to be deprecated.2. Sensitization of the recipientThe danger here chiefly relates to the Rh

antigen. A single transfusion of Rh-positiveblood given to an Rh-negative recipient mayresult in sensitization to this antigen and, even ifno obvious immediate harm results, may be apotential source of danger with subsequenttransfusions.

In an Rh-negative woman, such sensitization byRh positive blood is particularly to be avoidedsince indirectly it may have a devastating effect onfuture foetuses should she marry an Rh-positivemale. Not only are her infants more likely to beafflicted by haemolytic disease-and of a severetype-but the first child, which usually escapeswhen the immunization is due to pregnancy aloneis frequently diseased and may not survive. It is

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

June 1956 BREWER: The Hazards of Blood Transfusion 283

thus extremely important to ascertain prior totransfusion the Rh-group of any female patientin or before the child-bearing period of life, andonly to administer Rh-negative blood should sheprove to be in this category. Taking any chanceon this amounts to negligence. The counsel ofperfection, of course, is to Rh group all recipientsand to use only homologous Rh blood.

Repeated transfusions carry an enhanced risk ofsensitisation and the production of antibodies tosome of the rarer antigens. Certain patients suchas those with acquired haemolytic anaemia areespecially liable. As already indicated an indirectCoombs' test should be included in the directmatching in these cases.

3. Infected bloodInfected blood is a hazard of stored blood, since

freshly drawn blood is bactericidal and with afresh blood transfusion the blood is given beforesufficient time can elapse to allow of seriousmultiplication of any contaminating organisms.The bactericidal property of fresh blood, however,rapidly diminishes with refrigeration and storage,and stored blood becomes a good culture medium.It may even be to advantage to avoid refrigerationof freshly drawn blood for a few hours aftercollection and so enable it to overcome possiblestray contaminants.

Contamination may occur in many ways-fromthe skin of the donor at the time of bleeding (evenblood cultures taken under ideal conditions have asmall but definite contamination rate with skinorganisms), from the use of imperfectly sterilizedbottles, collecting sets or anticoagulant diluent,dependent on inefficient autoclaving, and by theintroduction of organisms when the bottle of bloodis opened for the preparation of ' packed cells.'The use of a pilot bottle has obviated the necessityof entering the main bottle to obtain a blood samplefor the direct matching test and has eliminated atleast one possible source of contamination. Anyinfection, however introduced, is likely in storedblood to result in a heavy bacterial growth if theblood is allowed to remain at room temperature fora few hours. Relating to this, the practice ofhaving several bottles of blood, which have beendirect matched for a particular patient, standingby in the theatre during an operation until theyare required, is to be condemned. Blood inreserve should be kept in the refrigerator until justbefore use.The transfusion of even a small volume of

grossly infected blood can produce a conditionakin to severe protein shock, with circulatoryfailure and rapid death. Such blood, dependenton the type of organism present, may or may notshow to naked-eye inspection haemolysis or a

pigment change, either of which should raisesuspicion. There may also be a faint smell ofhydrogen sulphide on opening the bottle. Atypicalcoliform bacilli capable of multiplying at refrigera-tor temperature have proved to be the mostsignificant organisms. When present in largenumbers an infecting organism may be de-monstrable in a direct smear made from theresidual blood in the bottle or from the heart'sblood of the patient post-mortem. Having inmind the risks to sterility to which stored bloodmay be exposed, it is gratifying that fatalities fromthis hazard are so few, but it is possible that minordegrees of bacterial contamination are responsiblefor some febrile reactions.

4. Circulatory overloadingThis is related not to the properties of the

transfused blood but to the total volume given andthe speed of administration. It is a transfusiondanger when the condition of the myocardium isimpaired as the result of severe chronic anaemia,cachexia or coronary disease. In such circum-stances any considerable increase in venouspressure may precipitate cardiac failure, and whentransfusion is indicated, the aim must be tointroduce red cells without producing any ap-preciable rise in venous pressure. Overloading inaddition to raising venous pressure results in anincrease in the amount of blood in the pulmonarycirculation, and a diminution in vital capacity.Prophylaxis consists in giving the blood at a veryslow drip rate (0.5-I c.c. per pound of body-weight per hour), in limiting the volume at onetransfusion to less than 500 c.c., in the use ofpacked cells which, compared with whole blood,achieve the same rise in haemoglobin level for halfthe volume and finally in warming the patient.This last procedure, by causing dilatation of theblood vessels in the skin and limbs, tends todiminish the amount of blood in the pulmonarycirculation, where at least part of the increasedblood volume resulting from too rapid a trans-fusion is normally accommodated. It also helpsto avoid a rigor which would put an additionalburden on the heart. Symptoms and signs whichgive warning of overloading are-tightness in thechest, a dry cough, dyspnoea, cyanosis, engorge-ment of the cervical veins, and an increased pulserate. Their appearance, if unheeded, is followedby pulmonary oedema with moist sounds at thelung bases and then rapid circulatory failure.Treatment consists in stopping the transfusion,propping the patient up and giving atropin and acardiac stimulant by hypodermic injection.

5. Air embolismAbout io c.c. of air introduced into the circula-

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

284 POSTGRADUATE MEDICAL JOURNAL June I956

tion of a sick person may prove fatal. Such anaccident may be due either to faulty technique orto faulty apparatus. The risk is greatest whenpositive air pressure, e.g. via a Higginson'ssyringe is employed to drive blood into theconstricted veins of a patient with oligaemic shock,or when an arterial transfusion is being carried out.A procedure of this nature should be in the handsof a responsible person lest the bottle becomeempty unnoticed.

Leaks in perished rubber tubing of the standardadministration set may allow air to be sucked intothe drip chamber if the screw clip controlling therate of flow is placed above the chamber. Toavoid this, the clip should be located as low downas possible. Other precautions against air em-bolism are ensuring that air is driven out of thetubing of the administration set by running bloodthrough -it before commencing a transfusion(the needle end of the set being held on a levelwith the bottle), and also changing to anotherbottle before the preceding one has quite emptiedso that the tubing is kept full of blood.

6. Transmissible diseasesThere are three diseases of practical importance

which may be transmitted-hepatitis, malaria andsyphilis. The danger of virus hepatitis has beenchiefly associated with the transfusion of pooledhuman plasma or serum. The incubation periodis long-80-I20 days, and the severity veryvariable. There may be mild transient jaundicewith little inconvenience or massive hepaticnecrosis with a fatal termination. In the past theincidence following the transfusion of these bloodproducts was as high as io per cent., but a re-duction in the number of donors contributing to apool to 8- o has greatly lowered the frequency andthe current incidence is about i per cent. Even sothis is still much higher than that of any otherinfection transmitted by transfusion and there ismuch to be said for the use of a plasma substitute(dextran is the most physiological of those avail-able) in place of plasma or serum until someeffective method of killing the virus is discovered.The virus survives drying and neither irradiationwith ultra-violet light nor treatment of the plasmaor serum with nitrogen mustard-two lines oferadication attempted-appear capable of achiev-ing this. The risk with whole blood is smallexcept with repeated transfusions or when a singlelarge volume transfusion entailing many donors isgiven (the hazard of hepatitis from o bottles ofblood approximates to I of ' small pool' plasmaor serum).Some years ago it was customary for prophylaxis

to exclude a volunteer from giving blood for aperiod of I2 months after a bout of hepatitis.

Recent work has demonstrated that the virus maypersist in the blood in a ' carrier' state for manyyears after the original attack, and it is nowconsidered safer not to accept any prospectivedonor giving a history of infective hepatitis, noranyone who has been a contact with the diseasewithin six months. Unrecognized sub-clinicalgrades of mild jaundice in donors are undoubtedlya potent source of infection. There is as yet noeasily performed laboratory test capable ofrevealing the presence of the virus.The problem of malaria is one mainly relating

to malarial endemic centres but it is of somesignificance in this country with the numerousex-Service men who have been in such areas andwith the present ease and facilities for foreigntravel. Observations bearing on this hazard arethat malarial parasites can survive in stored bloodrefrigerated at 2°-4°C throughout its period ofusage (i.e. up to 3 weeks from the date of collec-tion); also, that even when a donor has not residedin a malarious country for a very long time, hisblood may still harbour parasites, particularly inthe case of a quartan infection. With the efficacyof modern chemotherapy, the chances of anindividual transmitting this disease who has hadmalaria five years or more previously, and whohas not been exposed to the infection since, norhad any febrile bouts that could be attributable toit is extremely small.

Transfusion syphilis is now in Great Britain avery remote risk and still receding. Factorscontributing to this are the low incidence of thedisease, the efficacy of modern treatment andthe inability of the spirochaeta pallida to survivefor more than 72 hours in blood stored at 2-4°C.The limitations of a serological test, such as aWasserman reaction, as a safeguard with donorblood must be appreciated; the late primary orearly secondary stages of syphilis, times when theserological test may be negative or at least doubtful,are the most infectious periods, whereas thechronic tertiary in whom the result is stronglypositive is probably quite innocuous.

7. Febrile and allergic reactionsFebrile reactions. It is difficult to assess the true

incidence of febrile reactions associated with bloodtransfusion because figures quoted usually giveno information about the frequency with whichthe temperature of the patient was recorded, norover what period after the transfusion. In thewriter's experience if the temperature is takenhalf hourly during a transfusion and for at least 2hours afterwards few febrile reactions will beoverlooked and the incidence of fever up to ioo°Fis approximately 15 per cent. About 5 per cent.of patients have a greater rise which may be

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

June 1956 BREWER: The Hazards of Blood Transfusion 285

accompanied by a rigor. When a febrile reactiondevelops it usually does so within 90 minutes of theend of a rapid transfusion or during the course of aslow transfusion. Although the fever may extendto the next and following days it is most unlikelyfor a febrile reaction of this nature to commenceon the day after the transfusion. In such a case afurther cause must be sought.

Sterile compatible blood can produce a febrilereaction but there is no doubt that the chief causeis the presence in the anticoagulant fluid or theapparatus of foreign protein and polysaccharides-derived from dead bacteria, disintegrated or intact,and from the products of their growth. Theincidence of febrile reactions can be reduced to alow incidence by adopting scrupulous care in thepreparation of anticoagulant solutions and in thecleaning of used apparatus as follows-

I. The distilled water should be prepared inall-glass or specially designed stills.

2. All solutions must be prepared from freshlydistilled water (previous few hours) and notfrom water distilled on the previous day.

3. The solutions should be autoclaved as soonas they have been prepared.

4. The still itself and the bottles into which thesolutions are to be put must be rinsed withfreshly distilled water.

5. Used or new rubber tubing and glasswaremust be cleansed thoroughly so as toremove all debris and they should befinally rinsed in distilled water and thendried rapidly.

The introduction of expendable sets which arediscarded after use would be an advantage.The risk of febrile reactions is also lessened by(a) Giving the transfusion slowly e.g. a pint in

3-4 hours.(b) Keeping the patient warm, or even hot,

throughout the transfusion-e.g. wellwrapped, hot water bottles, windowsclosed to exclude draughts. This warm-ing up as previously mentioned alsoreduces the risk of circulatory overloading.

(c) In a nervous or already febrile patient theadministration of an opiate and/or anantihistamine drug before commencingthe transfusion.

A special point bearing on febrile reactions ininfants is worth emphasizing. It is that infantsseem incapable of shivering. They often, how-ever, with a febrile reaction exhibit pallor and havea cold skin, also accompanied by a refusal to feedand by some diarrhoea.

Allergic reactions. These are recognised by theappearance of a rash, swelling of mucous mem-branes, urticarial weals, itching and at timesnausea and vomiting. Asthma is rare and may

be difficult to distinguish from the results ofcirculatory overloading. Occasionally there maybe a rise of temperature. In most patients there isa history of some allergic affection. Thesereactions are usually associated with the donorplasma component of the transfusion and are dueto the release of histamine acting on the receptor'scells. Severe cases resemble shock and the patientshows marked apprehension. Prophylaxis in aknown allergic recipient may be achieved by thegiving intramuscularly of one of the antihistaminedrugs-antozoline hydrochloride (antistin) ioo mg.or promethazine hydrochloride (phenergan) 25mg. just before the transfusion. The addition ofchlorprophenpyridamine maleate (piriton) io mg.to the transfusion bottle immediately prior toadministration of the blood has also provedprotective. Packed cells (in which the amount ofplasma is minimal) or saline washed cells may beused with success for patients known to be allergicto plasma. For treatment-adrenalin hydro-chloride (I-iooo), Io minims, or ephedrinehydrochloride, 3 grain, by hypodermic injectionare indicated. Sensitivity can be passively trans-ferred from an allergic donor to a patient, parti-cularly if the donor's allergy is active at the timeof bleeding. Many specific examples are nowon record. The moral is to reject as blood donorsthose who give a history of hypersensitivity.8. Miscellaneous hazards

Transfusion thrombophlebitis is a not infrequentcomplication of prolonged blood transfusions butmore especially of saline and glucose infusions.Its incidence seems to be very variable in hospitalsin different parts of the country. The conditioncan prove painful to the patient and also seriouslyinterfere with intravenous post-operative fluidtherapy. The red rubber tubing used in thestandard transfusion administration sets has beensuspect and certainly latex rubber tubing appearsless liable in the hands of some operators to causetrouble. It is possible that a substance noxious tothe vein wall is gradually extracted from the redrubber tubing by the fluid passing through it-but this is by no means certain. Other factorssuch as rough handling of the vein may play a part.It is probably a wise precaution to employ a latextubing set for a transfusion likely to exceed I8hours and in the case of a saline and/or glucoseinfusion to change the needle to another vein afterit has run for this period.

Periodic blood transfusions may be life-sustaining for patients with a refractory, chronichaemolytic, or aplastic anaemia, but the necessity ofsuch regular administration of blood is liable intime to result in haemosiderosis. Since everypint of blood carries into the patient's body

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from

286 POSTGRADUATE MEDICAL JOURNAL June I956approximately 250 mg. of iron and since there isno pathway for excretion of iron, the giving of largenumbers of blood transfusions gradually leads tothe deposition in various organs of the body ofconsiderable quantities of iron which may in timeinterfere with their functions. Haemosiderosismay affect the bone marrow, myocardium, kidneysand adrenal glands. The liver may become cir-rhotic or diabetes may result from siderosis ofthe pancreas.

In conclusion every blood transfusion, even in

experienced hands, carries certain hazards and it isimportant that before a clinician recommends thisparticular therapy he should carefully weigh up foreach individual patient the advantages to be gainedand balance them against the potential dangersand ill effects.

ReferencesBull, G.M., Joekes,A.M., and Lowe K.G. (I949), Lancet, 2. 220,Discombe, G. (1955), " Blood Transfusion", Heinemann, LondonKeynes, G., (1949), " Blood Transfusion", Wright, Bristol.Mollison, P.L. (I95I), " Blood Transfusion in Clinical Medicine".

Blackwell, Oxford.

PULMONARY TUBERCULOSIS(Postgraduate Medical Journal)

Price 3s. lOd. post free

INTRODUCTION COLLAPSE THERAPY (MEDICAL) INMaurice Davidson, D.M., F.R.C.P. PULMONARY TUBERCULOSIS

F. H. Scadding, M.D., M.R.C.P.THE GENERAL MANAGEMENT OF THE SURGICAL COLLAPSE THERAPY INPHTHISICAL PATIENT FROM THE PULMONARY TBERCULOSISPHYSICIAN'S STANDPOINT J. R. BelcherL. E. Houghton, M.D. J. R. Belcher

RESECTION FOR PULMONARYTHE PRESENT POSITION OF CHEMO- TUBERCULOSISTHERAPYINTUBERCULOSIS W. P. Cleland, M.R.C.P., F.R.C.S.John C. Roberts, M.D., M.R.C.P. MODERN METHODS OF DIAGNOSIS IN

PULMONARY TUBERCULOSIS:THE PROBLEM OF TUBERCULOSIS IN CLINICAL, BACTERIOLOGICAL ANDCHILDHOOD RADIOLOGICALA. Margaret Macpherson, M.D., F.R.C.P. J. R. Bignall, M.D., M.R.C.P.

Published by

THE FELLOWSHIP OF POSTGRADUATE MEDICINE60, Portland Place, London, W.1

Index for 1955 is now ready. Have your Journalsbound up into a volume NOW. See advertise-

ment on page 296 for details.

copyright. on F

ebruary 26, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.32.368.279 on 1 June 1956. D

ownloaded from