- Email: [email protected]
DEFECTIVE GAS-TRANSPORT FUNCTION OF STORED RED BLOOD-CELLS
119
fashioning an ileostomy particular care must be exercised with regard to its situation ; the waist-line, bony prominences, operation scars, and the umbilicus In
should be avoided. Patients must be taught the detailed management of their ileostomy before leaving hospital. The adhesive types of appliances are definitely superior to the non-adhesive types and should therefore be used whenever possible.
the oxygen and carbon-dioxide We dissociation curves and their interrelationship. and that take in the here place oxygen report changes carbon-dioxide dissociation curves of blood after storage under standard blood-bank conditions, and alterations in the oxygen dissociation curve of the recipient of such
investigated by studying
blood.
Summary A report is presented of a survey of 52 patients with ileostemies undertaken to determine the degree of disability caused by an ileostomy, and how these patients managed in their homes. The large majority led happy and useful lives and were in gainful employment, mostly in their former
occupations. Certain points
in the
technique
of
fashioning
an
ileos-
tomy and in its subsequent management both by the surgeon and by the patient are discussed, and it is emphasised that, if the disability is to be kept to a minimum, the ileostomy must be correctly sited and constructed, and an adhesive appliance worn. Not all
patients, however, are
suitable for such an appliance ; so the management of non-adhesive types is also discussed. The main complications to which these patients are liable are mentioned, with particular reference to the danger of rapid and severe fluid and electrolyte loss from obstruction. We
greatly indebted to the surgeons of St. Mark’s for allowing us to review their cases, and to Dr. Cuthbert Dukes and Mr. Hedley Atkins for their help. We acknowledge with thanks a grant received from the board of governors of the Hammersmith, West London, and St. Mark’s Hospitals. We are also grateful to Miss Sylvia Treadgold and Miss M. J. Waldron of the department of medical illustration of Guy’s Hospital for the drawings. are
Hospital
REFERENCES
Brooke, B. N. (1952) Lancet, ii, 102. Cattell, R. B. (1939) Sur. Clin. N. Amer. 19, 629. — (1948) Gastro-enterology, 10, 63. Cave, H. W. (1945) Surg. Clin. N. Amer. 25, 301. Counsell, P. B., Goligher, J. C. (1952) Lancet, ii, 1045. Dragstedt, L. R., Dack, G. M., Kirsner, J. B. (1941) Ann.-Surg. 114, 653. Lahey, F. H. (1951) Ibid, 133, 726. Lancet (1948) ii, 306. McKittrick, L. S., Moore, F. D. (1949) J. Amer. med. Ass. 139, 201. Warren, R., McKittrick, L. S. (1951) Surg. Gynec. Obstet. 93, 555. Wells, C. A. (1952) Brit. J. Surg. 39, 309.
DEFECTIVE GAS-TRANSPORT FUNCTION OF STORED RED BLOOD-CELLS M.D. Athens MEDICINE,
UNIVERSITY OF
THESSALONIKI ;
VISITING RESEARCH-WORKER
M.B. MEDICAL
THE
A. C. KENNEDY Glasg., M.R.C.P.E., F.R.F.P.S.
REGISTRAR,
ROYAL
Preparation of Blood Samples for In-vitro Studies Samples of stored blood were obtained from blood supplied by the local blood-transfusion service for routine transfusions, stored
by
us
from blood obtained from volunteers and in universal containers.
or
The anticoagulant medium (A.C.D.) was composed of disodium citrate 2 g., dextrose 3 g., and water to 100 ml. The proportion was 100 ml. of A.C.D. to 440 ml. of blood, and the temperature of storage was 4°C. For samples of fresh blood the anticoagulant medium, of pH 7-4, was 4% sodium fluoride in 0-2% heparin solution, in a proportion of 1 drop to 3 ml. of blood (Riley et al. 1946). Blood stored with this anticoagulant at 4°C undergoes no change in plasma pH (pHs) up to twenty days. For studies of acidified fresh blood the heparin-fluoride anticoagulant was acidified with a normal solution of lactic acid in saline solution. Laked blood buffered solutions were prepared by the methods of Brooks (1935) and Darling and Roughton (1942). The plasma pH of the blood samples was partially restored by substituting fresh plasma for the acid plasma ; or, to bring the plasma pH further to normal, in some cases the stored cells were washed with fresh plasma before the substitution. In these manoeuvres care was taken to avoid exposure to atmospheric air.
Oxygen
and Carbon-dioxide Dissociation Curves
Tonometers of 300 ml. capacity were filled at the desired gas tensions (Austin et al. 1922, Dill 1928) by a modification (Van Slyke et al. 1923) of the apparatus described by Austin et al. The blood samples of 3 ml. were equilibrated at the gas tensions of the tonometer by mechanical rotation in a waterbath at 37°C for thirty minutes. Four points were determined for most of the oxygen dissociation curves (at partial pressure of oxygen of 20, 40, and 100 mm. Hg and atmospheric air), and in a few cases additional determinations were made at 60 and 80 mm. Hg. The partial pressure of carbon dioxide in the tonometers was 40 mm. Hg. The oxygen capacity was obtained by- rotation in tonometers containing atmospheric air at 37°C (Roughton et al. 1944). Two points were determined on each carbon-dioxide curve from oxygenated blood, and one point from reduced blood. The blood-gas analyses were made in the Van Slyke manometric apparatus (Peters and Van Slyke 1932) with 1-ml. blood samples and the lactic acid, ferricyanide, and urea reagent (King et al. 1948, Rappaport 1949). Correction of the position of the oxygen dissociation curve to standard plasma pH 7-4 was effected by use of the equation
log p02 plasma pH1944).
-0’48
D. J. VALTIS LECTURER IN
Material and Methods
INFIRMARY,
GLASGOW
major purpose of transfusion of whole blood, compared with the transfusion of plasma or plasma products, is to provide red cells which can perform their normal function of transporting oxygen to, and removing carbon dioxide from, the tissues. Despite the obvious importance of the gas-transport function of stored blood, the only observations in this field are, so far as we are aware, reports on the oxygen capacity of stored blood (Belk et al. 1939, Scarborough and Thompson 1940, Denstedt et al. 1941). The oxygen capacity, however, does not provide, as has been erroneously stated (Rapoport 1947), an indication of the rate of reaction between haemoglobin and oxygen on which depends the release-of oxygen to the tissues. This can be satisfactorily
(Keys
et
al.
1936, Aste-Salazar and
The value 0-48 has been confirmed by us for Hurtado fresh blood of plasma pH down to 7-0, but at lower plasma pH levels we found the value to be 0-46. Correction of the oxygen dissociation curve for the presence of carbon monoxide (Riley et al. 1946) was required only where storage was for more than
thirty days. The plasma pH
of true plasma, from blood equilibrated at 37°C with arterial gas tensions, was determined both with
glass electrode standardised with buffers ofpH 6-5 and 7-4 and by the Henderson-Hasselbalch equation, using for pK’ and the solubility coefficient of carbon dioxide in serum the values given by Dill et al. (1937, 1940) and Van Slyke et al. (1928). We have used this solubility coefficient for citrated plasma ; but any error so produced in the in-vitro studies is small because the solubility coefficient of carbon dioxide in a citrate solution is lower than that of carbon dioxide in water (Van Slyke et al. 1928), and the value is very near that for the plasma. Indeed, the calculated values for plasma pH, obtained at 37°C, agreed closely ( 002) with those obtained ’with a glass electrode at room-temperature, after the temperature difference has been allowed for. To avoid small errors resulting from variations in room-temperature and the loss of carbon dioxide to the atmosphere during the glass-electrode determinations the calculated pH values are used.
a
°
120 The total carbon dioxide of the plasma (total CO2)s, was determined directly from plasma obtained from blood saturated with oxygen at a carbon dioxide partial pressure of 40 mm. Hg. The total carbon dioxide of oxygenated blood (total CO2)O, and of reduced blood (total CO2)r, was also determined directly. Spectrophotometry for carboxyhæmoglobin (Heilmeyer 1943) and methæmoglobin and sulphsemoglobin (Evelyn and Malloy 1938) was done with a’Unicam S.P. 600’ spectro-
TABLE
II—OXYGEN, CARBON-DIOXIDE, AND CARBON-MONOXIDE CONTENT OF STORED
CITRATED BLOOD
photometer. Studies transfused blood, prepared in the standard manner by the West of Scotland Blood Transfusion Service, had been stored from half an hour to twenty days. For the plasma transfusion we used plasma freshly separated from citrated blood stored for twenty days. The observations were made on ten patients, all severely anaemic, receiving fourteen transfusions ; the speed of transfusion was a pint in one or two hours. Oxygen dissociation curves and determinations of blood and plasma carbon dioxide were done in all the cases immediately before and immediately after transfusion and in some of the cases from half an hour to twenty days later. The time-consuming nature of these investigations prevented more frequent estimations. The data for the curves were obtained from venous blood, taken under the standard conditions described for gas analysis, using the heparin fluoride anticoagulant.
Transfusion The
Results IN-VITRO STUDIES
and carbon-dioxide dissociation curves were 100 samples of blood stored for up to ninety days. As the changes observed were fully developed after twenty days, most of the results presented are from this, the clinically important, period. The results are shown in tables I-III and figs. 1-4. The position of the oxygen dissociation curve is indicated in table I and figs. 1 and 2 by the partial pressure of oxygen at which 50% of the haemoglobin is oxygenated.
Oxygen
made
on
over
TABLE I-RESULTS
*Using the formula 5 from table iv of Austin et al. (1922) and 0-15 as solubility coefficient for CO. at that temperature.
Effect of Hæmolysis, Dilution, and pH Change in Fresh Normal Blood (table i) The values obtained by the above-mentioned tech. niques in normal blood samples agree with those given by other workers (Dill 1928, Keys et al. 1936, Aste. Salazar and Hurtado 1944). Since dilution and, some. times, slight hoemolysis are present in stored blood, we studied slightly hæmolysed samples of fresh normal blood diluted with plasma to the degree present in stored blood, and found that neither dilution (Richards and Strauss 1927) nor slight haemolysis had an effect on the oxygen (Total CO2) r-o, dissociation curve or on the quotient ,..
which indicates the amount of carbon dioxide released from the blood for each volume % of oxygen saturation of haemoglobin (figs. 1 and 4). In fresh acidified heparin. ised blood the oxygen dissociation curve and the quotient
(Total CO2)
r—o are
.,in
keeping
with
the
low
plasma pH ; after partial restoration of the plasma pH with fresh plasma there is no abnormality (figs. 2 and 4). Blood Stored in A.C.D. Medium (tables i and II) The oxygen dissociation curve (figs. 1-3) shows a shift to the left, which is progressive with storage, and the
IS
(MEAN VALUES) OF GAS STUDIES IN FRESH BLOOD, HÆMOLYSED AND DILUTED FRESH ACIDIFIED FRESH BLOOD, AND BLOOD STORED IN VARIOUS ANTICOAGULANTS
BLOOD,
Blood sample
Fresh blood
..........
Fresh blood slightly heemolysed and diluted Acidified fresh blood ..... Partial restoration of pH with plasma ..
Stored in A.C.D. for 0-24 hr..... Partial restoration of pH with plasma Stored in A.C.D. for 7 days Partial restoration of pH with plasma Stored in acid heparin for 7 days Partial restoration of pH with plasma ..
Stored in trisodium citrate for 7
days
Stored in heparin fluoride for 7 days Stored in A.C.D. for 20 days ..‘ Partial restoration of pH with plasma
..
..
..
.
,
Stored plasma plus fresh red cells.. Partial restoration of pH with plasma
.
Stored in
A.C.D.
for 2
i
mos.
*Corrected for carbon monoxide.
(Total C02)r, total carbon dioxide of reduced blood at CO2 partial pressure of 40 nun. Hg. (Total C02)o, total carbon dioxide of oxygenated blood at a CO2 partial pressure of 40 mm. Hg. , relation between the difference (Total CO,,)r -(Total COz)o and oxyhæmoglobin. (Total CO2), total carbon dioxide of true plasma from oxygenated blood at COz partial pressure of 40 mm. Hg. Blood plasma pH, plasma pH determined in true plasma from blood saturated with oxygen when PCO2=40 mm. Hg. PO2 for Hb = HbO2, partial pressure of oxygen at which haemoglobin saturation with oxygen is 50 % at the plasma pH of the recipient and at standard plasma pH of 7-4.
121
is
abnormalities
are
(fig. 4). The not present in changes are very slight after storage for one day, pro- blood stored for nounc4-after seven days, and even more pronounced, seven days with heparinalthough not to the degree expected from the further the plasma pH fall (6-8-6-6), after twenty days. These fluoride anticoagulant (figs. abnormalities persist after partial restoration of the Studies fresh after calculated correction 1-4). with plasma ; plasma pH the sevenbeyond of the position of the oxygen dissociation curve to standay period were dard plasma pH 7-4 the shift to the left is more obvious impossible in (figs. 2 and 3). After two months’ storage the shift of the oxygen heparinised dissociation curve to the left is still greater, but after samples because correction for the presence of carboxyhæmoglobin the of gross haemoposition of the curve does not differ essentially from lysis. that of citrated blood stored twenty days (fig. 2). Laked Blood SoluThe addition of citrated plasma, derived from citrated tions (table III) blood stored twenty days, to fresh red cells does not The position and shape of the produce the abnormalities (figs. 1, 2, and 4).
quotient
decreased
TABLE III-OXYGEN DISSOCIATION
CURVES
IN
LAKED
BLOOD
SOLUTIONS FROM FRESH BLOOD AND STORED CITRATED BLOOD
Fig. 3-Oxygen dissociation
curves at plasma pH 74 : a, blood stored in A.C.D. for twenty days ; b, blood stored in A.C.D. for
of
blood stored in trisodium in acidified heparin for seven daysd, blood stored in A.C.D. for one day; interrupted line, normal fresh blood.
seven
oxygen dissociation curve from buffered solutions
days ;c,
citrate
or
laked
blood from citrated blood stored twenty days, is the The same as that given by solutions from fresh blood. curve from citrated blood stored for two months lies slightly to the left of the normal position, probably owing to the presence of carboxyhæmoglobin.
prepared
TRANSFUSION STUDIES
Table 11 shows that the carbon-dioxide and carbonmonoxide content of citrated blood become progressively increased with storage, and this is accompanied by a fall in the oxygen content and oxyhæmoglobin capacity and a rise in the calculated partial pressure of carbon dioxide from 20 to 65 mm. Hg. Blood Stored in Other Anticoagulants (table 1) Blood samples stored for seven days either with trisodium citrate or with acid heparin show basically the same abnormalities as the blood stored for the same period in A.C.D. medium but in lesser degree ; the
-
.
Fig. I-Relation between position of oxygen dissociation curve and Plasma pH of fresh blood and of blood stored in various anticoagulants. *
0
.. s + e
X
a 13
The results are presented in tables iv and v and figs. 5-7. The transfusion of 1 pint of blood thirty minutes after its withdrawal into the A.c.D. medium (table iv, case 1) did not significantly shift the recipient’s oxygen dissociation curve immediately after transfusion, whereas the transfusion of 2 pints of blood twenty hours after withdrawal into the A.c.D. medium (case 2) shifted the oxygen dissociation curve slightly to the left (fig. 5). In all the patients who received 2 or 3 pints of blood stored for seven days or more, the recipient’s oxygen dissociation curve was substantially shifted to the left immediately after the transfusion ; the shift was greater in those receiving 3 pints. In cases 3, 5, 7, 9, 9a, and 10 oxygen dissociation curves were plotted from half an
Fresh normal blood Fresh blood haemotysed and diluted Stored in A.C.D. 1/2-24 hr. Stored in A.C.D. days Stored in acid heparin 7 days Stored in trisodium citrate 7 days Stored in heparin fluoride 7 days Stored in A.C.D. 20 days Plasma from stored blood plus fresh red cells
Fig. 2-Position of
oxygen dissociation
curve at
plasma pH
7.4 of-fresh
blood, acidified fresh blood, and blood stored in various anticoagulants for different periods of time. Note change of time scale. .
0 A A
+ e x
0
Fresh normal blood Acidified fresh blood Stored in A.C.D. Stored in A.C.D. after CO correction Stored in acid heparin Stored in trisodium citrate Stored in heparin fluoride Plasma from stored blood plus fresh red cells.
122
2-pint transfusion, when the oxygen dissociation curve still slightly to the left ; the shift of the curve after the second transfusion lasted several days. The degree of shift of the recipient’s oxygen dissocia. tion curve did not appear to increase with time of storage of the transfused blood beyond seven days. In all the cases receiving stored citrated blood the difference between the carbon dioxide of the recipient’s reduced and oxygenated blood was altered, but the relationship between the carbon-dioxide difference and the oxyhsemoglobin capacity remained unchanged. The plasma pH of the recipient after transfusion of stored citrated blood was in no case significantly altered. The transfusion of 600 ml. of plasma, derived from 3 pints of citrated blood stored twenty days, produced no significant change in the recipient’s oxygen dissociation curve or plasma pH (table v). Gas analysis of the recipient’s blood after transfusion of stored blood did not reveal any significant amount of carbon monoxide, and spectrophotometry for methæmoglobin, carboxyhsemoglobin, or other abnormal pigments The parenteral administration of was also negative. ascorbic acid 1000 mg. or of methylene-blue 2 mg. per kg. of body-weight immediately after the transfusion of 2 pints of stored blood did not prevent the shift of the oxygen dissociation curve to the left (table v).
a
was
Fig.
between
and
4—Relation between 4-Relation
and of blood stored in various
plasma pH of fresh blood anticoagulants.
* Fresh normal blood Fresh blood h2emolysed and diluted * Stored in A.C.D. 1/2-24 hr. &Dgr; Stored in A.C.D. 7 days + Stored in acid heparin 7 days e Stored in trisodium citrate 7 days x Stored in heparin fluoride 7 days * Stored in A.C.D. 20 days a Plasma from stored blood plus fresh red cells o
hour up to six hours after transfusion, when the persistent shift to the left was still considerable (fig. 6) ; in cases 5, 6, and 9 the curve was still slightly abnormal at twentyfour hours. In case 10, who received 3 pints of blood stored one or two weeks, the curve was still definitely abnormal twenty-four hours after transfusion (fig. 7). In case 9a a 3-pint transfusion was given four days after
Discussion SIGNIFICANCE OF CHANGES
The in-vitro results show that the oxygen dissociation of stored citrated blood is shifted to the left, suggesting that such blood after transfusion will be, at least temporarily, incapable of releasing as large a curve
TABLE IV-EFFECT OF TRANSFUSION OF STORED CITRATED BLOOD A
Time of estimation tr in relation to transfusion b
f1
Immediately before Immediately after
Immediately before Immediately after Immediately before Immediately after 25 min. after 24 hr. after
I
Immediately before
Immediately after . Immediately before Immediately after 620hr.hr.after after
before Immediately Immediately after
24 hr. after 8 days after
Immediately before 30 min. after 16 hr. after
I
Immediately before I
Immediately after
Immediately before Immediately after 1 hr. after 24 hr. after : days after
i
Immediately before Immediately after I 1 hr. after
24 hr. after 3 days after 8
daysafter
20 days after Immediately before Immediately after
424hr. after
hr. after
123 TABLE V-TRANSFUSION OF PLASMA FROM BLOOD STORED FOR 20 DAYS AND EFFECT OF ADMINISTRATION OF METHYLENE-BLUE AND ASCORBIC ACID
Administration of ascorbic acid
volume of oxygen to the tissues as normal blood. The clinical studies confirm that after transfusion of stored citrated blood the recipient’s oxygen dissociation curve is shifted to the left, actually to a degree greater than might be expected from the in-vitro findings. The magnitude of this shift is proportional to the amount and age of the transfused blood. The shift is present immediately after transfusion and, in the case of a transfusion of 2 or more pints stored a few days, remains for several hours. The curve may remain slightly abnormal for several days if several pints of stored blood are transfused. The clinical significance of the shift to the left is shown by the effect of a transfusion of 3 pints of citrated blood stored for one or two weeks to a patient with haemoglobin 35% (table iv, case 10). Although the haemoglobin was increased to 55%, the shift of the oxygen dissociation curve to the left (fig. 7) means that after transfusion the patient’s blood could not deliver to the tissues as much oxygen as it did before. Thus, given a normal oxygen partial pressure of 40 mm. Hg in the tissues, this patient’s blood before transfusion released to the tissues about 40% of the oxygen carried (about 3 vols. of oxygen %). For several hours after transfusion, given the same oxygen partial pressure in the tissues, the blood could release only 20% of the oxygen carried (about 2 vols. of oxygen %). These immediate post-transfusion hours are often very critical for the seriously ill patient. The body has, of course, many compensatory mechanisms, but in the critically ill patient it is conceivable that these may not operate
rapidly enough
or
to
The carbon-dioxide alterations in stored citrated blood of less importance than the oxygen changes and indeed may be beneficial. For each volume per cent. of oxygen saturation haemoglobin normally releases 0-35-0-4 volume- of carbon dioxide at plasma pH 7-4, and 0-3 volume at plasma pH 7-3. In stored citrated blood with plasma pH 7°3this value is 0,2. This alteration suggests that the recipient of stored citrated blood will be able to release the normal volume of carbon dioxide in the lungs only if the tension of carbon dioxide in the tissues is increased. Such an increase will tend to lower the plasma pH and at the same time to stimulate the respiratory centre. Both these secondary effects would exert a beneficial influence on oxygen uptake in the lungs and release in the tissues. The clinical studies, however, show that the carbon-dioxide abnormality observed in vitro, beneficial or otherwise, is not present after the transfusion of stored citrated blood. are
CAUSATION OF CHANGES
The changes observed in seven-day-stored citrated blood are not present in blood stored in heparin for the same period at pH 7-4, or in acidified fresh blood ;
a
sufficient extent. It was several days before the curve returned completely to its pre-transfusion position, but the smaller percentage of oxygen released-e.g., after six hours-was more than Fig. 5-Position at plasma pH 7.4 of recipient’s oxygen dissociation curve before (interrupted line) compensated by the higher and after (continuous line) transfusion of 2 pints of citrated blood stored twenty hours. haemoglobin level.
Fig. 6-Position
at
plasma pH 74 of recipient’s
curve before (interrupted line), immediately after (a), and six hours after (b), transfusion of 2 pints of citrated blood stored seven days.
oxygen
dissociation
124
they are present, but in considerably lesser degree, in blood stored for seven days in acid heparin or in trisodium citrate. These observations suggest that storage, acid reaction, and citrate must all be present together for the full development of the abnormalities. Changes in the plasma of stored citrated blood, resulting from red-cell metabolism or from abnormal breakdown products of haemoglobin, do not appear to be responsible for the abnormality of the oxygen dissociation curve, for the shift is not produced by the addition of plasma derived from stored citrated blood to fresh red cells in vitro, or by the transfusion of such plasma. An alkalosis would cause a shift of the oxygen dissociation curve to the left, but we confirmed the finding of Wurmser et al. (1942) and Loutit et al. (1943) that there is little or no alteration in the recipient’s plasma pH after transfusion of stored citrated blood. Moreover correction of the position of the recipient’s oxygen dissociation curve to the standard plasma pH does not diminish the shift of the curve to the left. Carboxyhæmoglobin (Stadie and Martin 1925) and methæmoglobin (Darling and Roughton 1942) shift the oxygen dissociation curve to theleft. Appreciableamounts of carboxy-
haemoglobin are present in blood stored in heparin at
room-
temperature (Sj ostran d 1951b), and the optimum conditions for conversion of
haemoglobin to methaemoglobin within the red cell are
storage
under low oxygen and carbonhigh Fig. 7-Position at plasma pH 7.4 of recipient’s dioxide presoxygen dissociation curve before (interrupted
line), immediately after (a), four hours after sure (Brooks (b), and twenty-four hours after (c), transfusion 1935), condiof 3 pints of citrated blood stored from seven totions existing fourteen days. in stored cit-
rated blood has also reported an increased alveolar concentration of carbon monoxide in the recipient after transfusion. We were unable, however, either by gas analysis or by spectrophotometry, to find any significant amounts of carboxyhaemoglobin, methæmoglobin, or other abnormal pigment-e.g., sulphæmoglobin—in citrated blood stored for thirty days or in the recipient’s blood after transfusion. The absence of methaemoglobin is confirmed indirectly by the fact that the oxygen capacity of citrated blood stored seven days is slightly greater than that of fresh citrated blood (table II), possibly indicating the conversion of inactive pigment (Ammundsen 1941) to an active form, and by the failure of methylene-blue and of ascorbic acid to prevent the shift of the curve to the left. A chemical change in haemoglobin structure or the presence of other chemical substances-e.g., glutathione (Litarczek et al. 1931)--cannot be responsible for the changes described since they are not present in laked blood solutions prepared from stored citrated blood
(table II).
Sjostrand (1951a)
(table III). In view of our observation that the patient’s oxygen dissociation curve is almost completely restored to normal a few hours after transfusion of stored citrated blood, it is significant that earlier workers (Mollison and Young
1941, Harris 1941, Maizels and Paterson 1940, Maizels 1943, 1949) have shown that the physical and chemical in the stored red cell largely disappear after the cells have been some hours in the recipient’s circulation. That the shift of the oxygen dissociation curve is due to an alteration in the physical, chemical, and electrolytic relations between cell and plasma produced by storage is supported by our own studies on the causes and prevention of the abnormality (Valtis and Kennedy
changes
- 1953). The small shift of the curve, which may persist several is perhaps due to the admixture of the cells of the anæmic recipient with the donor cells. The oxygen dissociation curve in anaemia lies to the right of normal (Richards and Strauss 1927, Henderson 1928, Isac et al. 1938), and the addition of cells from a normal person would be expected to produce an oxygen dissocia. tion curve intermediate in position between the normal and the anaemic curves, the exact position depending on the proportion of donor and recipient cells. This view is supported by the case which showed a progressive shift of the oxygen dissociation curve to the left with succes. sive transfusions. Meyer and DuBois (1916) reported a slight increase in the metabolic rate in pernicious anaemia,, and Tompkins et al. (1919) found a decrease in the rate after transfusion. The shift of the anaemic patient’s oxygen dissociation curve to the right explains the increased metabolic rate, and the shift to the left after transfusion the decrease.
days after transfusion,
Summary
of the oxygen and carbon-dioxide dissociation curves of blood stored in an acid-citrate-dextrose medium at 4°C has revealed undesirable effects of storage not apparent by the indices commonly used for assessing storage conditions. The oxygen dissociation curve was shifted to the left, and the amount of carbon dioxide released for each volume per cent. of haemoglobin saturation with oxygen was reduced. These changes were progressive with storage. The oxygen dissociation curves of patients after transfusion with citrated blood stored seven days or more was substantially shifted to the left immediately after transfusion, and this effect lasted several hours. The magnitude and duration of the shift were propor. tional to the volume and length of time of storage of the transfused blood. As a result of this shift the anaemic recipient’s blood may be unable, for a few hours after transfusion, to release as much oxygen as it did before. For the full development of the abnormalities storage, acid reaction, and citrate are necessary. The changes are not due to abnormal pigments, such as carboxyhæmoglobin and methaemoglobin, or to alteration of the
Study
recipient’s plasma pH. We wish to thank Prof. L. J. Davis for his helpful advice and criticism throughout this study ; Dr. J. Wallace, of the West of Scotland Blood Transfusion Service, and Dr. J. C. Eaton, biochemist to the Royal Infirmary, Glasgow, for cooperation ; and Dr. A. G. Baikie for the spectrophotometric studies. One of us (D. J. V.) was enabled to undertake this work by scholarships from the E.C.A. for Greece and from the University of Thessaloniki. Much of the cost of apparatus and materials was provided by the Rankin Fund of the University of Glasgow. REFERENCES. Ammundsen, E. (1941) J. biol. Chem. 138, 563. Aste-Salazar, H., Hurtado, A. (1944) Amer. J. Physiol. 142, 733. Austin, J. H., Cullen, G. E., Hastings, A. B., McLean, F. C., Peters, J. P., Van Slyke, D. D. (1922) J. biol. Chem. 54, 121. Belk, W. P., Henry, N. W., Rosenstein, F. (1939) Amer. J. med. Sci. 198, 631. Brooks, J. (1935) Proc. roy. Soc. B, 118, 560. Darling, R. C., Roughton, F. J. W. (1942) Amer. J. Physiol. 137, 56. Denstedt, O. F., Osborne, D. E., Roche, M. N., Stansfield, H. (1941) Canad. med. Ass. J. 44, 448. Dill, D. B. (1928) In Henderson, L. J. Blood, a Study in General Physiology. New Haven, Conn. References continued at foot of next column
125
BACTERIOLOGY OF AIR AND DUST IN A MATERNITY HOSPITAL FRANK MARSH M.D. Lond., D.T.M. & H. PATHOLOGIST,
ST.
MARGARET’S HOSPITAL, EPPING,
ESSEX
HELEN E. RODWAY F.R.C.O.G. OBSTETRICIAN,
THORPE COOMBE MATERNITY
HOSPITAL,
The final investigation was made in the lying-in wards of the main hospital block during the nine months November, 1952, to July, 1953. Some bacteriological counts of the ward air were made at this time, but the chief object then was to find out whether the disinfectant measures in use were effective in preventing and controlling the spread of pathogenic organisms among the babies nursed in the hospital wards. Results of First Bacterial Dust in Two 4-bed
WALTHAMSTOW. LONDON
BETWEEN May, 1950, and July, 1953, we tested from time to time the air and dust in the lying-in wards of a maternity hospital where the babies were nursed in cots beside their mothers’ beds. We did this to find the type of organisms harboured in the ward dust and the variations in counts at different times of the day and night, and we tried to reduce the number of bacteria by using, as a ward routine, dust-suppressive measures, such as damp dusting and sweeping and the oiling of linoleumcovered and wooden floors. We tried also to clean the air by spraying an aerosol from a ’Phantomyser ’ and by using an electrostatic dust-precipitator. Finally, we tried to lessen the dispersal of bacteria-laden dust from bedclothes, by disinfecting blankets, mattresses, and
The average bacterial
Colony Counts of Lying-in Wards
Air and
counts for plates exposed wards during May and June, 1950, are shown in table i. For 18 plates exposed in a lower-floor ward the average counts were 36 colonies from midnight until 5 A.M. ; 187 from 5 A.M. until 5 P.M. ; and 35 from 5 P.M. to midnight. For 20 plates exposed in an upper-floor ward the average counts for these three periods of the day were 27, 132, and
in two 4-bed
TABLE
I-AVERAGE
PLATES
colony
lying-in
EXPOSED
COUNTS
BACTERIAL
OF
4-BED WARDS AND JUNE, 1950
IN TWO
HOSPITAL DURING MAY
COLONIES
FROM
OF A MATERNITY
pillows. In the first
plates of
an
part of the investigation bacteriological exposed in two 4-bed wards for periods hour, at different times throughout the day and were
night during May and June, 1950, and again from October, 1950, to January, 1951. Bacterial counts were made after forty-eight hours’ incubation. Later, from September, 1951, to the end of January, 1952, the investigation was continued in single wards of a segregation block, where the results of the disinfection of blankets and bedding were tested. Plates were exposed during this investigation hourly from 1 P.M. to 4 P.M., so as to include an active bedmaking period between two quiet periods.
57 colonies respectively. Separate plate counts showed that in quiet periods the bacterial colonies were few, but during ward activity, particularly bedmaking, colony counts rose to a high level, sometimes increasing 40-50 times. The bacterial colonies consisted chiefly of sarcinse, Micrococcus pharyngeus, Staphylococcus albus, a coliform bacillus, and Bacillus subtilis. A few unidentified fungi were also grown. The plates occasionally showed scanty colonies of Staph. pyogenes, a pneumococcus, and a non-haemolytic streptococcus ; and twice during these observations one or two colonies of a hæmolytic streptococcus were seen.
Daly, C., Forbes, W. H. (1937) J. biol. Chem. 117, 569. Graybiel, A., Hurtado, A., Taquini, A. C. (1940) Z. Alterforsch. 2, 20. Evelyn, K. A., Malloy, H. T. (1938) J. biol. Chem. 126, 655. Harris, J. E. (1941) Ibid, 141, 579. Heilmeyer, L. (1943) Spectrophotometry in Medicine. London. Henderson, L. J. (1928) Blood. New Haven, Conn. Isac, C., Matthes, K., Yamanaka, T. (1938) Arch. exp. Path. Pharmak. 189, 615. Keys, A., Hall, F. G., Barron, E. S. G. (1936) Amer. J. Physiol. 115, 292. King, E. J., Gilchrist, M., Wootton, I. D. P., O’Brien, J. R. P., Jope, H. M., Quelch, P. E., Peterson, J. M., Strangeways, D. H., Ramsay, W. N. M. (1948) Lancet, i, 478. Litarczek, G., Aubert, H., Cosmulesco, J. (1931) C.R. Soc. Biol. Paris, 106, 973. Loutit, J. F., Mollison, P. L., Young, I. M. (1943) Quart. J. exp. Physiol. 32, 183. Maizels, M. (1943) Ibid, p. 143. — (1949) J. Physiol. 108, 247. — Paterson, J. H. (1940) Lancet, ii, 417. Meyer, A. L., DuBois, E. F. (1916) Arch. intern. Med. 17, 965. Mollison, P. L., Young, I. M. (1941) Brit. med. J. ii, 797. Peters, J. P., Van Slyke, D. D. (1932) Quantitative Clinical Chemistry. Baltimore. Rapoport, S. (1947) J. clin. Invest. 26, 591. Rappaport, F. (1949) Rapid Microchemical Methods for Blood and —
—
C S F Examinations. New York. W. jun., Strauss, M. L. (1927) J. clin. Invest. 4, 105. Riley, R. L., Lilienthal, J. L. jun., Proemmel, D. D., Franke, R. E. (1946) Ibid, 25, 139. Roughton, F. J. W., Darling, R. C., Root, W. S. (1944) Amer. J. Physiol. 142, 708. Scarborough, H., Thompson, J. C. (1940) Edinb. med. J. 47, 567. Sjöstrand, T. (1951a) Acta physiol. scand. 22, 137. — (1951b) Ibid, p. 142. Stadie, W. C., Martin, K. A. (1925) J. clin. Invest. 2, 77. Tompkins, E. H., Brittingham, H. H., Drinker, C. K. (1919) Arch. intern. Med. 23, 441. Valtis, D. J., Kennedy, A. C. (1953) Glasg. med. J. 34, 521. Van Slyke, D. D., Sendroy, J. jun., Hastings, A. B., Neill, J. M. (1928) J. biol. Chem. 78, 765. — Wu, H., McLean, F. C. (1923) Ibid, 56, 765. Wurmser, R., Filitti-Wurmser, S., Briault, R. (1942) Rev. canad.
Richards, D.
Biol. 1, 372.
Table
the bacterial counts from plates during a winter period (October, 1950, to January, 1951). For 89 plates exposed in the lower-floor ward the average counts were 53 for the quiet period from midnight until 5 A.M. ; 182 during the day-time from 5 A.M. to 5 P.M. ; and 91 from 5 P.M. to midnight. For 71 plates exposed in the upperfloor ward the average counts for the three periods In these tests the were 26, 211, and 105 respectively. separate plates showed that the evening counts in winter were higher than those of the spring period, probably because ward ventilation was often unsatisfactory during the winter evenings. Table lib shows the average bacterial colony counts after intermittent treatment of the air with an aerosol during the same winter period (October, 1950, to January, 1951). For 21 plates exposed in the lower-floor 4bed ward the average counts for the three periods of the
exposed
na
gives
at intervals
TABLE IIA-AVERAGE
PLATES
EXPOSED
COUNTS DURING
OF BACTERIAL COLONIES FROM PERIOD FROM OCTOBER TO
A
JANUARY, 1951
TABLE IIB—AVERAGE COUNTS OF BACTERIAL COLONIES AFTER TREATMENT OF AIR WITH AEROSOL DURING A PERIOD FROM
OCTOBER TO
JANUARY, 1951
fashioning an ileostomy particular care must be exercised with regard to its situation ; the waist-line, bony prominences, operation scars, and the umbilicus In
should be avoided. Patients must be taught the detailed management of their ileostomy before leaving hospital. The adhesive types of appliances are definitely superior to the non-adhesive types and should therefore be used whenever possible.
the oxygen and carbon-dioxide We dissociation curves and their interrelationship. and that take in the here place oxygen report changes carbon-dioxide dissociation curves of blood after storage under standard blood-bank conditions, and alterations in the oxygen dissociation curve of the recipient of such
investigated by studying
blood.
Summary A report is presented of a survey of 52 patients with ileostemies undertaken to determine the degree of disability caused by an ileostomy, and how these patients managed in their homes. The large majority led happy and useful lives and were in gainful employment, mostly in their former
occupations. Certain points
in the
technique
of
fashioning
an
ileos-
tomy and in its subsequent management both by the surgeon and by the patient are discussed, and it is emphasised that, if the disability is to be kept to a minimum, the ileostomy must be correctly sited and constructed, and an adhesive appliance worn. Not all
patients, however, are
suitable for such an appliance ; so the management of non-adhesive types is also discussed. The main complications to which these patients are liable are mentioned, with particular reference to the danger of rapid and severe fluid and electrolyte loss from obstruction. We
greatly indebted to the surgeons of St. Mark’s for allowing us to review their cases, and to Dr. Cuthbert Dukes and Mr. Hedley Atkins for their help. We acknowledge with thanks a grant received from the board of governors of the Hammersmith, West London, and St. Mark’s Hospitals. We are also grateful to Miss Sylvia Treadgold and Miss M. J. Waldron of the department of medical illustration of Guy’s Hospital for the drawings. are
Hospital
REFERENCES
Brooke, B. N. (1952) Lancet, ii, 102. Cattell, R. B. (1939) Sur. Clin. N. Amer. 19, 629. — (1948) Gastro-enterology, 10, 63. Cave, H. W. (1945) Surg. Clin. N. Amer. 25, 301. Counsell, P. B., Goligher, J. C. (1952) Lancet, ii, 1045. Dragstedt, L. R., Dack, G. M., Kirsner, J. B. (1941) Ann.-Surg. 114, 653. Lahey, F. H. (1951) Ibid, 133, 726. Lancet (1948) ii, 306. McKittrick, L. S., Moore, F. D. (1949) J. Amer. med. Ass. 139, 201. Warren, R., McKittrick, L. S. (1951) Surg. Gynec. Obstet. 93, 555. Wells, C. A. (1952) Brit. J. Surg. 39, 309.
DEFECTIVE GAS-TRANSPORT FUNCTION OF STORED RED BLOOD-CELLS M.D. Athens MEDICINE,
UNIVERSITY OF
THESSALONIKI ;
VISITING RESEARCH-WORKER
M.B. MEDICAL
THE
A. C. KENNEDY Glasg., M.R.C.P.E., F.R.F.P.S.
REGISTRAR,
ROYAL
Preparation of Blood Samples for In-vitro Studies Samples of stored blood were obtained from blood supplied by the local blood-transfusion service for routine transfusions, stored
by
us
from blood obtained from volunteers and in universal containers.
or
The anticoagulant medium (A.C.D.) was composed of disodium citrate 2 g., dextrose 3 g., and water to 100 ml. The proportion was 100 ml. of A.C.D. to 440 ml. of blood, and the temperature of storage was 4°C. For samples of fresh blood the anticoagulant medium, of pH 7-4, was 4% sodium fluoride in 0-2% heparin solution, in a proportion of 1 drop to 3 ml. of blood (Riley et al. 1946). Blood stored with this anticoagulant at 4°C undergoes no change in plasma pH (pHs) up to twenty days. For studies of acidified fresh blood the heparin-fluoride anticoagulant was acidified with a normal solution of lactic acid in saline solution. Laked blood buffered solutions were prepared by the methods of Brooks (1935) and Darling and Roughton (1942). The plasma pH of the blood samples was partially restored by substituting fresh plasma for the acid plasma ; or, to bring the plasma pH further to normal, in some cases the stored cells were washed with fresh plasma before the substitution. In these manoeuvres care was taken to avoid exposure to atmospheric air.
Oxygen
and Carbon-dioxide Dissociation Curves
Tonometers of 300 ml. capacity were filled at the desired gas tensions (Austin et al. 1922, Dill 1928) by a modification (Van Slyke et al. 1923) of the apparatus described by Austin et al. The blood samples of 3 ml. were equilibrated at the gas tensions of the tonometer by mechanical rotation in a waterbath at 37°C for thirty minutes. Four points were determined for most of the oxygen dissociation curves (at partial pressure of oxygen of 20, 40, and 100 mm. Hg and atmospheric air), and in a few cases additional determinations were made at 60 and 80 mm. Hg. The partial pressure of carbon dioxide in the tonometers was 40 mm. Hg. The oxygen capacity was obtained by- rotation in tonometers containing atmospheric air at 37°C (Roughton et al. 1944). Two points were determined on each carbon-dioxide curve from oxygenated blood, and one point from reduced blood. The blood-gas analyses were made in the Van Slyke manometric apparatus (Peters and Van Slyke 1932) with 1-ml. blood samples and the lactic acid, ferricyanide, and urea reagent (King et al. 1948, Rappaport 1949). Correction of the position of the oxygen dissociation curve to standard plasma pH 7-4 was effected by use of the equation
log p02 plasma pH1944).
-0’48
D. J. VALTIS LECTURER IN
Material and Methods
INFIRMARY,
GLASGOW
major purpose of transfusion of whole blood, compared with the transfusion of plasma or plasma products, is to provide red cells which can perform their normal function of transporting oxygen to, and removing carbon dioxide from, the tissues. Despite the obvious importance of the gas-transport function of stored blood, the only observations in this field are, so far as we are aware, reports on the oxygen capacity of stored blood (Belk et al. 1939, Scarborough and Thompson 1940, Denstedt et al. 1941). The oxygen capacity, however, does not provide, as has been erroneously stated (Rapoport 1947), an indication of the rate of reaction between haemoglobin and oxygen on which depends the release-of oxygen to the tissues. This can be satisfactorily
(Keys
et
al.
1936, Aste-Salazar and
The value 0-48 has been confirmed by us for Hurtado fresh blood of plasma pH down to 7-0, but at lower plasma pH levels we found the value to be 0-46. Correction of the oxygen dissociation curve for the presence of carbon monoxide (Riley et al. 1946) was required only where storage was for more than
thirty days. The plasma pH
of true plasma, from blood equilibrated at 37°C with arterial gas tensions, was determined both with
glass electrode standardised with buffers ofpH 6-5 and 7-4 and by the Henderson-Hasselbalch equation, using for pK’ and the solubility coefficient of carbon dioxide in serum the values given by Dill et al. (1937, 1940) and Van Slyke et al. (1928). We have used this solubility coefficient for citrated plasma ; but any error so produced in the in-vitro studies is small because the solubility coefficient of carbon dioxide in a citrate solution is lower than that of carbon dioxide in water (Van Slyke et al. 1928), and the value is very near that for the plasma. Indeed, the calculated values for plasma pH, obtained at 37°C, agreed closely ( 002) with those obtained ’with a glass electrode at room-temperature, after the temperature difference has been allowed for. To avoid small errors resulting from variations in room-temperature and the loss of carbon dioxide to the atmosphere during the glass-electrode determinations the calculated pH values are used.
a
°
120 The total carbon dioxide of the plasma (total CO2)s, was determined directly from plasma obtained from blood saturated with oxygen at a carbon dioxide partial pressure of 40 mm. Hg. The total carbon dioxide of oxygenated blood (total CO2)O, and of reduced blood (total CO2)r, was also determined directly. Spectrophotometry for carboxyhæmoglobin (Heilmeyer 1943) and methæmoglobin and sulphsemoglobin (Evelyn and Malloy 1938) was done with a’Unicam S.P. 600’ spectro-
TABLE
II—OXYGEN, CARBON-DIOXIDE, AND CARBON-MONOXIDE CONTENT OF STORED
CITRATED BLOOD
photometer. Studies transfused blood, prepared in the standard manner by the West of Scotland Blood Transfusion Service, had been stored from half an hour to twenty days. For the plasma transfusion we used plasma freshly separated from citrated blood stored for twenty days. The observations were made on ten patients, all severely anaemic, receiving fourteen transfusions ; the speed of transfusion was a pint in one or two hours. Oxygen dissociation curves and determinations of blood and plasma carbon dioxide were done in all the cases immediately before and immediately after transfusion and in some of the cases from half an hour to twenty days later. The time-consuming nature of these investigations prevented more frequent estimations. The data for the curves were obtained from venous blood, taken under the standard conditions described for gas analysis, using the heparin fluoride anticoagulant.
Transfusion The
Results IN-VITRO STUDIES
and carbon-dioxide dissociation curves were 100 samples of blood stored for up to ninety days. As the changes observed were fully developed after twenty days, most of the results presented are from this, the clinically important, period. The results are shown in tables I-III and figs. 1-4. The position of the oxygen dissociation curve is indicated in table I and figs. 1 and 2 by the partial pressure of oxygen at which 50% of the haemoglobin is oxygenated.
Oxygen
made
on
over
TABLE I-RESULTS
*Using the formula 5 from table iv of Austin et al. (1922) and 0-15 as solubility coefficient for CO. at that temperature.
Effect of Hæmolysis, Dilution, and pH Change in Fresh Normal Blood (table i) The values obtained by the above-mentioned tech. niques in normal blood samples agree with those given by other workers (Dill 1928, Keys et al. 1936, Aste. Salazar and Hurtado 1944). Since dilution and, some. times, slight hoemolysis are present in stored blood, we studied slightly hæmolysed samples of fresh normal blood diluted with plasma to the degree present in stored blood, and found that neither dilution (Richards and Strauss 1927) nor slight haemolysis had an effect on the oxygen (Total CO2) r-o, dissociation curve or on the quotient ,..
which indicates the amount of carbon dioxide released from the blood for each volume % of oxygen saturation of haemoglobin (figs. 1 and 4). In fresh acidified heparin. ised blood the oxygen dissociation curve and the quotient
(Total CO2)
r—o are
.,in
keeping
with
the
low
plasma pH ; after partial restoration of the plasma pH with fresh plasma there is no abnormality (figs. 2 and 4). Blood Stored in A.C.D. Medium (tables i and II) The oxygen dissociation curve (figs. 1-3) shows a shift to the left, which is progressive with storage, and the
IS
(MEAN VALUES) OF GAS STUDIES IN FRESH BLOOD, HÆMOLYSED AND DILUTED FRESH ACIDIFIED FRESH BLOOD, AND BLOOD STORED IN VARIOUS ANTICOAGULANTS
BLOOD,
Blood sample
Fresh blood
..........
Fresh blood slightly heemolysed and diluted Acidified fresh blood ..... Partial restoration of pH with plasma ..
Stored in A.C.D. for 0-24 hr..... Partial restoration of pH with plasma Stored in A.C.D. for 7 days Partial restoration of pH with plasma Stored in acid heparin for 7 days Partial restoration of pH with plasma ..
Stored in trisodium citrate for 7
days
Stored in heparin fluoride for 7 days Stored in A.C.D. for 20 days ..‘ Partial restoration of pH with plasma
..
..
..
.
,
Stored plasma plus fresh red cells.. Partial restoration of pH with plasma
.
Stored in
A.C.D.
for 2
i
mos.
*Corrected for carbon monoxide.
(Total C02)r, total carbon dioxide of reduced blood at CO2 partial pressure of 40 nun. Hg. (Total C02)o, total carbon dioxide of oxygenated blood at a CO2 partial pressure of 40 mm. Hg. , relation between the difference (Total CO,,)r -(Total COz)o and oxyhæmoglobin. (Total CO2), total carbon dioxide of true plasma from oxygenated blood at COz partial pressure of 40 mm. Hg. Blood plasma pH, plasma pH determined in true plasma from blood saturated with oxygen when PCO2=40 mm. Hg. PO2 for Hb = HbO2, partial pressure of oxygen at which haemoglobin saturation with oxygen is 50 % at the plasma pH of the recipient and at standard plasma pH of 7-4.
121
is
abnormalities
are
(fig. 4). The not present in changes are very slight after storage for one day, pro- blood stored for nounc4-after seven days, and even more pronounced, seven days with heparinalthough not to the degree expected from the further the plasma pH fall (6-8-6-6), after twenty days. These fluoride anticoagulant (figs. abnormalities persist after partial restoration of the Studies fresh after calculated correction 1-4). with plasma ; plasma pH the sevenbeyond of the position of the oxygen dissociation curve to standay period were dard plasma pH 7-4 the shift to the left is more obvious impossible in (figs. 2 and 3). After two months’ storage the shift of the oxygen heparinised dissociation curve to the left is still greater, but after samples because correction for the presence of carboxyhæmoglobin the of gross haemoposition of the curve does not differ essentially from lysis. that of citrated blood stored twenty days (fig. 2). Laked Blood SoluThe addition of citrated plasma, derived from citrated tions (table III) blood stored twenty days, to fresh red cells does not The position and shape of the produce the abnormalities (figs. 1, 2, and 4).
quotient
decreased
TABLE III-OXYGEN DISSOCIATION
CURVES
IN
LAKED
BLOOD
SOLUTIONS FROM FRESH BLOOD AND STORED CITRATED BLOOD
Fig. 3-Oxygen dissociation
curves at plasma pH 74 : a, blood stored in A.C.D. for twenty days ; b, blood stored in A.C.D. for
of
blood stored in trisodium in acidified heparin for seven daysd, blood stored in A.C.D. for one day; interrupted line, normal fresh blood.
seven
oxygen dissociation curve from buffered solutions
days ;c,
citrate
or
laked
blood from citrated blood stored twenty days, is the The same as that given by solutions from fresh blood. curve from citrated blood stored for two months lies slightly to the left of the normal position, probably owing to the presence of carboxyhæmoglobin.
prepared
TRANSFUSION STUDIES
Table 11 shows that the carbon-dioxide and carbonmonoxide content of citrated blood become progressively increased with storage, and this is accompanied by a fall in the oxygen content and oxyhæmoglobin capacity and a rise in the calculated partial pressure of carbon dioxide from 20 to 65 mm. Hg. Blood Stored in Other Anticoagulants (table 1) Blood samples stored for seven days either with trisodium citrate or with acid heparin show basically the same abnormalities as the blood stored for the same period in A.C.D. medium but in lesser degree ; the
-
.
Fig. I-Relation between position of oxygen dissociation curve and Plasma pH of fresh blood and of blood stored in various anticoagulants. *
0
.. s + e
X
a 13
The results are presented in tables iv and v and figs. 5-7. The transfusion of 1 pint of blood thirty minutes after its withdrawal into the A.c.D. medium (table iv, case 1) did not significantly shift the recipient’s oxygen dissociation curve immediately after transfusion, whereas the transfusion of 2 pints of blood twenty hours after withdrawal into the A.c.D. medium (case 2) shifted the oxygen dissociation curve slightly to the left (fig. 5). In all the patients who received 2 or 3 pints of blood stored for seven days or more, the recipient’s oxygen dissociation curve was substantially shifted to the left immediately after the transfusion ; the shift was greater in those receiving 3 pints. In cases 3, 5, 7, 9, 9a, and 10 oxygen dissociation curves were plotted from half an
Fresh normal blood Fresh blood haemotysed and diluted Stored in A.C.D. 1/2-24 hr. Stored in A.C.D. days Stored in acid heparin 7 days Stored in trisodium citrate 7 days Stored in heparin fluoride 7 days Stored in A.C.D. 20 days Plasma from stored blood plus fresh red cells
Fig. 2-Position of
oxygen dissociation
curve at
plasma pH
7.4 of-fresh
blood, acidified fresh blood, and blood stored in various anticoagulants for different periods of time. Note change of time scale. .
0 A A
+ e x
0
Fresh normal blood Acidified fresh blood Stored in A.C.D. Stored in A.C.D. after CO correction Stored in acid heparin Stored in trisodium citrate Stored in heparin fluoride Plasma from stored blood plus fresh red cells.
122
2-pint transfusion, when the oxygen dissociation curve still slightly to the left ; the shift of the curve after the second transfusion lasted several days. The degree of shift of the recipient’s oxygen dissocia. tion curve did not appear to increase with time of storage of the transfused blood beyond seven days. In all the cases receiving stored citrated blood the difference between the carbon dioxide of the recipient’s reduced and oxygenated blood was altered, but the relationship between the carbon-dioxide difference and the oxyhsemoglobin capacity remained unchanged. The plasma pH of the recipient after transfusion of stored citrated blood was in no case significantly altered. The transfusion of 600 ml. of plasma, derived from 3 pints of citrated blood stored twenty days, produced no significant change in the recipient’s oxygen dissociation curve or plasma pH (table v). Gas analysis of the recipient’s blood after transfusion of stored blood did not reveal any significant amount of carbon monoxide, and spectrophotometry for methæmoglobin, carboxyhsemoglobin, or other abnormal pigments The parenteral administration of was also negative. ascorbic acid 1000 mg. or of methylene-blue 2 mg. per kg. of body-weight immediately after the transfusion of 2 pints of stored blood did not prevent the shift of the oxygen dissociation curve to the left (table v).
a
was
Fig.
between
and
4—Relation between 4-Relation
and of blood stored in various
plasma pH of fresh blood anticoagulants.
* Fresh normal blood Fresh blood h2emolysed and diluted * Stored in A.C.D. 1/2-24 hr. &Dgr; Stored in A.C.D. 7 days + Stored in acid heparin 7 days e Stored in trisodium citrate 7 days x Stored in heparin fluoride 7 days * Stored in A.C.D. 20 days a Plasma from stored blood plus fresh red cells o
hour up to six hours after transfusion, when the persistent shift to the left was still considerable (fig. 6) ; in cases 5, 6, and 9 the curve was still slightly abnormal at twentyfour hours. In case 10, who received 3 pints of blood stored one or two weeks, the curve was still definitely abnormal twenty-four hours after transfusion (fig. 7). In case 9a a 3-pint transfusion was given four days after
Discussion SIGNIFICANCE OF CHANGES
The in-vitro results show that the oxygen dissociation of stored citrated blood is shifted to the left, suggesting that such blood after transfusion will be, at least temporarily, incapable of releasing as large a curve
TABLE IV-EFFECT OF TRANSFUSION OF STORED CITRATED BLOOD A
Time of estimation tr in relation to transfusion b
f1
Immediately before Immediately after
Immediately before Immediately after Immediately before Immediately after 25 min. after 24 hr. after
I
Immediately before
Immediately after . Immediately before Immediately after 620hr.hr.after after
before Immediately Immediately after
24 hr. after 8 days after
Immediately before 30 min. after 16 hr. after
I
Immediately before I
Immediately after
Immediately before Immediately after 1 hr. after 24 hr. after : days after
i
Immediately before Immediately after I 1 hr. after
24 hr. after 3 days after 8
daysafter
20 days after Immediately before Immediately after
424hr. after
hr. after
123 TABLE V-TRANSFUSION OF PLASMA FROM BLOOD STORED FOR 20 DAYS AND EFFECT OF ADMINISTRATION OF METHYLENE-BLUE AND ASCORBIC ACID
Administration of ascorbic acid
volume of oxygen to the tissues as normal blood. The clinical studies confirm that after transfusion of stored citrated blood the recipient’s oxygen dissociation curve is shifted to the left, actually to a degree greater than might be expected from the in-vitro findings. The magnitude of this shift is proportional to the amount and age of the transfused blood. The shift is present immediately after transfusion and, in the case of a transfusion of 2 or more pints stored a few days, remains for several hours. The curve may remain slightly abnormal for several days if several pints of stored blood are transfused. The clinical significance of the shift to the left is shown by the effect of a transfusion of 3 pints of citrated blood stored for one or two weeks to a patient with haemoglobin 35% (table iv, case 10). Although the haemoglobin was increased to 55%, the shift of the oxygen dissociation curve to the left (fig. 7) means that after transfusion the patient’s blood could not deliver to the tissues as much oxygen as it did before. Thus, given a normal oxygen partial pressure of 40 mm. Hg in the tissues, this patient’s blood before transfusion released to the tissues about 40% of the oxygen carried (about 3 vols. of oxygen %). For several hours after transfusion, given the same oxygen partial pressure in the tissues, the blood could release only 20% of the oxygen carried (about 2 vols. of oxygen %). These immediate post-transfusion hours are often very critical for the seriously ill patient. The body has, of course, many compensatory mechanisms, but in the critically ill patient it is conceivable that these may not operate
rapidly enough
or
to
The carbon-dioxide alterations in stored citrated blood of less importance than the oxygen changes and indeed may be beneficial. For each volume per cent. of oxygen saturation haemoglobin normally releases 0-35-0-4 volume- of carbon dioxide at plasma pH 7-4, and 0-3 volume at plasma pH 7-3. In stored citrated blood with plasma pH 7°3this value is 0,2. This alteration suggests that the recipient of stored citrated blood will be able to release the normal volume of carbon dioxide in the lungs only if the tension of carbon dioxide in the tissues is increased. Such an increase will tend to lower the plasma pH and at the same time to stimulate the respiratory centre. Both these secondary effects would exert a beneficial influence on oxygen uptake in the lungs and release in the tissues. The clinical studies, however, show that the carbon-dioxide abnormality observed in vitro, beneficial or otherwise, is not present after the transfusion of stored citrated blood. are
CAUSATION OF CHANGES
The changes observed in seven-day-stored citrated blood are not present in blood stored in heparin for the same period at pH 7-4, or in acidified fresh blood ;
a
sufficient extent. It was several days before the curve returned completely to its pre-transfusion position, but the smaller percentage of oxygen released-e.g., after six hours-was more than Fig. 5-Position at plasma pH 7.4 of recipient’s oxygen dissociation curve before (interrupted line) compensated by the higher and after (continuous line) transfusion of 2 pints of citrated blood stored twenty hours. haemoglobin level.
Fig. 6-Position
at
plasma pH 74 of recipient’s
curve before (interrupted line), immediately after (a), and six hours after (b), transfusion of 2 pints of citrated blood stored seven days.
oxygen
dissociation
124
they are present, but in considerably lesser degree, in blood stored for seven days in acid heparin or in trisodium citrate. These observations suggest that storage, acid reaction, and citrate must all be present together for the full development of the abnormalities. Changes in the plasma of stored citrated blood, resulting from red-cell metabolism or from abnormal breakdown products of haemoglobin, do not appear to be responsible for the abnormality of the oxygen dissociation curve, for the shift is not produced by the addition of plasma derived from stored citrated blood to fresh red cells in vitro, or by the transfusion of such plasma. An alkalosis would cause a shift of the oxygen dissociation curve to the left, but we confirmed the finding of Wurmser et al. (1942) and Loutit et al. (1943) that there is little or no alteration in the recipient’s plasma pH after transfusion of stored citrated blood. Moreover correction of the position of the recipient’s oxygen dissociation curve to the standard plasma pH does not diminish the shift of the curve to the left. Carboxyhæmoglobin (Stadie and Martin 1925) and methæmoglobin (Darling and Roughton 1942) shift the oxygen dissociation curve to theleft. Appreciableamounts of carboxy-
haemoglobin are present in blood stored in heparin at
room-
temperature (Sj ostran d 1951b), and the optimum conditions for conversion of
haemoglobin to methaemoglobin within the red cell are
storage
under low oxygen and carbonhigh Fig. 7-Position at plasma pH 7.4 of recipient’s dioxide presoxygen dissociation curve before (interrupted
line), immediately after (a), four hours after sure (Brooks (b), and twenty-four hours after (c), transfusion 1935), condiof 3 pints of citrated blood stored from seven totions existing fourteen days. in stored cit-
rated blood has also reported an increased alveolar concentration of carbon monoxide in the recipient after transfusion. We were unable, however, either by gas analysis or by spectrophotometry, to find any significant amounts of carboxyhaemoglobin, methæmoglobin, or other abnormal pigment-e.g., sulphæmoglobin—in citrated blood stored for thirty days or in the recipient’s blood after transfusion. The absence of methaemoglobin is confirmed indirectly by the fact that the oxygen capacity of citrated blood stored seven days is slightly greater than that of fresh citrated blood (table II), possibly indicating the conversion of inactive pigment (Ammundsen 1941) to an active form, and by the failure of methylene-blue and of ascorbic acid to prevent the shift of the curve to the left. A chemical change in haemoglobin structure or the presence of other chemical substances-e.g., glutathione (Litarczek et al. 1931)--cannot be responsible for the changes described since they are not present in laked blood solutions prepared from stored citrated blood
(table II).
Sjostrand (1951a)
(table III). In view of our observation that the patient’s oxygen dissociation curve is almost completely restored to normal a few hours after transfusion of stored citrated blood, it is significant that earlier workers (Mollison and Young
1941, Harris 1941, Maizels and Paterson 1940, Maizels 1943, 1949) have shown that the physical and chemical in the stored red cell largely disappear after the cells have been some hours in the recipient’s circulation. That the shift of the oxygen dissociation curve is due to an alteration in the physical, chemical, and electrolytic relations between cell and plasma produced by storage is supported by our own studies on the causes and prevention of the abnormality (Valtis and Kennedy
changes
- 1953). The small shift of the curve, which may persist several is perhaps due to the admixture of the cells of the anæmic recipient with the donor cells. The oxygen dissociation curve in anaemia lies to the right of normal (Richards and Strauss 1927, Henderson 1928, Isac et al. 1938), and the addition of cells from a normal person would be expected to produce an oxygen dissocia. tion curve intermediate in position between the normal and the anaemic curves, the exact position depending on the proportion of donor and recipient cells. This view is supported by the case which showed a progressive shift of the oxygen dissociation curve to the left with succes. sive transfusions. Meyer and DuBois (1916) reported a slight increase in the metabolic rate in pernicious anaemia,, and Tompkins et al. (1919) found a decrease in the rate after transfusion. The shift of the anaemic patient’s oxygen dissociation curve to the right explains the increased metabolic rate, and the shift to the left after transfusion the decrease.
days after transfusion,
Summary
of the oxygen and carbon-dioxide dissociation curves of blood stored in an acid-citrate-dextrose medium at 4°C has revealed undesirable effects of storage not apparent by the indices commonly used for assessing storage conditions. The oxygen dissociation curve was shifted to the left, and the amount of carbon dioxide released for each volume per cent. of haemoglobin saturation with oxygen was reduced. These changes were progressive with storage. The oxygen dissociation curves of patients after transfusion with citrated blood stored seven days or more was substantially shifted to the left immediately after transfusion, and this effect lasted several hours. The magnitude and duration of the shift were propor. tional to the volume and length of time of storage of the transfused blood. As a result of this shift the anaemic recipient’s blood may be unable, for a few hours after transfusion, to release as much oxygen as it did before. For the full development of the abnormalities storage, acid reaction, and citrate are necessary. The changes are not due to abnormal pigments, such as carboxyhæmoglobin and methaemoglobin, or to alteration of the
Study
recipient’s plasma pH. We wish to thank Prof. L. J. Davis for his helpful advice and criticism throughout this study ; Dr. J. Wallace, of the West of Scotland Blood Transfusion Service, and Dr. J. C. Eaton, biochemist to the Royal Infirmary, Glasgow, for cooperation ; and Dr. A. G. Baikie for the spectrophotometric studies. One of us (D. J. V.) was enabled to undertake this work by scholarships from the E.C.A. for Greece and from the University of Thessaloniki. Much of the cost of apparatus and materials was provided by the Rankin Fund of the University of Glasgow. REFERENCES. Ammundsen, E. (1941) J. biol. Chem. 138, 563. Aste-Salazar, H., Hurtado, A. (1944) Amer. J. Physiol. 142, 733. Austin, J. H., Cullen, G. E., Hastings, A. B., McLean, F. C., Peters, J. P., Van Slyke, D. D. (1922) J. biol. Chem. 54, 121. Belk, W. P., Henry, N. W., Rosenstein, F. (1939) Amer. J. med. Sci. 198, 631. Brooks, J. (1935) Proc. roy. Soc. B, 118, 560. Darling, R. C., Roughton, F. J. W. (1942) Amer. J. Physiol. 137, 56. Denstedt, O. F., Osborne, D. E., Roche, M. N., Stansfield, H. (1941) Canad. med. Ass. J. 44, 448. Dill, D. B. (1928) In Henderson, L. J. Blood, a Study in General Physiology. New Haven, Conn. References continued at foot of next column
125
BACTERIOLOGY OF AIR AND DUST IN A MATERNITY HOSPITAL FRANK MARSH M.D. Lond., D.T.M. & H. PATHOLOGIST,
ST.
MARGARET’S HOSPITAL, EPPING,
ESSEX
HELEN E. RODWAY F.R.C.O.G. OBSTETRICIAN,
THORPE COOMBE MATERNITY
HOSPITAL,
The final investigation was made in the lying-in wards of the main hospital block during the nine months November, 1952, to July, 1953. Some bacteriological counts of the ward air were made at this time, but the chief object then was to find out whether the disinfectant measures in use were effective in preventing and controlling the spread of pathogenic organisms among the babies nursed in the hospital wards. Results of First Bacterial Dust in Two 4-bed
WALTHAMSTOW. LONDON
BETWEEN May, 1950, and July, 1953, we tested from time to time the air and dust in the lying-in wards of a maternity hospital where the babies were nursed in cots beside their mothers’ beds. We did this to find the type of organisms harboured in the ward dust and the variations in counts at different times of the day and night, and we tried to reduce the number of bacteria by using, as a ward routine, dust-suppressive measures, such as damp dusting and sweeping and the oiling of linoleumcovered and wooden floors. We tried also to clean the air by spraying an aerosol from a ’Phantomyser ’ and by using an electrostatic dust-precipitator. Finally, we tried to lessen the dispersal of bacteria-laden dust from bedclothes, by disinfecting blankets, mattresses, and
The average bacterial
Colony Counts of Lying-in Wards
Air and
counts for plates exposed wards during May and June, 1950, are shown in table i. For 18 plates exposed in a lower-floor ward the average counts were 36 colonies from midnight until 5 A.M. ; 187 from 5 A.M. until 5 P.M. ; and 35 from 5 P.M. to midnight. For 20 plates exposed in an upper-floor ward the average counts for these three periods of the day were 27, 132, and
in two 4-bed
TABLE
I-AVERAGE
PLATES
colony
lying-in
EXPOSED
COUNTS
BACTERIAL
OF
4-BED WARDS AND JUNE, 1950
IN TWO
HOSPITAL DURING MAY
COLONIES
FROM
OF A MATERNITY
pillows. In the first
plates of
an
part of the investigation bacteriological exposed in two 4-bed wards for periods hour, at different times throughout the day and were
night during May and June, 1950, and again from October, 1950, to January, 1951. Bacterial counts were made after forty-eight hours’ incubation. Later, from September, 1951, to the end of January, 1952, the investigation was continued in single wards of a segregation block, where the results of the disinfection of blankets and bedding were tested. Plates were exposed during this investigation hourly from 1 P.M. to 4 P.M., so as to include an active bedmaking period between two quiet periods.
57 colonies respectively. Separate plate counts showed that in quiet periods the bacterial colonies were few, but during ward activity, particularly bedmaking, colony counts rose to a high level, sometimes increasing 40-50 times. The bacterial colonies consisted chiefly of sarcinse, Micrococcus pharyngeus, Staphylococcus albus, a coliform bacillus, and Bacillus subtilis. A few unidentified fungi were also grown. The plates occasionally showed scanty colonies of Staph. pyogenes, a pneumococcus, and a non-haemolytic streptococcus ; and twice during these observations one or two colonies of a hæmolytic streptococcus were seen.
Daly, C., Forbes, W. H. (1937) J. biol. Chem. 117, 569. Graybiel, A., Hurtado, A., Taquini, A. C. (1940) Z. Alterforsch. 2, 20. Evelyn, K. A., Malloy, H. T. (1938) J. biol. Chem. 126, 655. Harris, J. E. (1941) Ibid, 141, 579. Heilmeyer, L. (1943) Spectrophotometry in Medicine. London. Henderson, L. J. (1928) Blood. New Haven, Conn. Isac, C., Matthes, K., Yamanaka, T. (1938) Arch. exp. Path. Pharmak. 189, 615. Keys, A., Hall, F. G., Barron, E. S. G. (1936) Amer. J. Physiol. 115, 292. King, E. J., Gilchrist, M., Wootton, I. D. P., O’Brien, J. R. P., Jope, H. M., Quelch, P. E., Peterson, J. M., Strangeways, D. H., Ramsay, W. N. M. (1948) Lancet, i, 478. Litarczek, G., Aubert, H., Cosmulesco, J. (1931) C.R. Soc. Biol. Paris, 106, 973. Loutit, J. F., Mollison, P. L., Young, I. M. (1943) Quart. J. exp. Physiol. 32, 183. Maizels, M. (1943) Ibid, p. 143. — (1949) J. Physiol. 108, 247. — Paterson, J. H. (1940) Lancet, ii, 417. Meyer, A. L., DuBois, E. F. (1916) Arch. intern. Med. 17, 965. Mollison, P. L., Young, I. M. (1941) Brit. med. J. ii, 797. Peters, J. P., Van Slyke, D. D. (1932) Quantitative Clinical Chemistry. Baltimore. Rapoport, S. (1947) J. clin. Invest. 26, 591. Rappaport, F. (1949) Rapid Microchemical Methods for Blood and —
—
C S F Examinations. New York. W. jun., Strauss, M. L. (1927) J. clin. Invest. 4, 105. Riley, R. L., Lilienthal, J. L. jun., Proemmel, D. D., Franke, R. E. (1946) Ibid, 25, 139. Roughton, F. J. W., Darling, R. C., Root, W. S. (1944) Amer. J. Physiol. 142, 708. Scarborough, H., Thompson, J. C. (1940) Edinb. med. J. 47, 567. Sjöstrand, T. (1951a) Acta physiol. scand. 22, 137. — (1951b) Ibid, p. 142. Stadie, W. C., Martin, K. A. (1925) J. clin. Invest. 2, 77. Tompkins, E. H., Brittingham, H. H., Drinker, C. K. (1919) Arch. intern. Med. 23, 441. Valtis, D. J., Kennedy, A. C. (1953) Glasg. med. J. 34, 521. Van Slyke, D. D., Sendroy, J. jun., Hastings, A. B., Neill, J. M. (1928) J. biol. Chem. 78, 765. — Wu, H., McLean, F. C. (1923) Ibid, 56, 765. Wurmser, R., Filitti-Wurmser, S., Briault, R. (1942) Rev. canad.
Richards, D.
Biol. 1, 372.
Table
the bacterial counts from plates during a winter period (October, 1950, to January, 1951). For 89 plates exposed in the lower-floor ward the average counts were 53 for the quiet period from midnight until 5 A.M. ; 182 during the day-time from 5 A.M. to 5 P.M. ; and 91 from 5 P.M. to midnight. For 71 plates exposed in the upperfloor ward the average counts for the three periods In these tests the were 26, 211, and 105 respectively. separate plates showed that the evening counts in winter were higher than those of the spring period, probably because ward ventilation was often unsatisfactory during the winter evenings. Table lib shows the average bacterial colony counts after intermittent treatment of the air with an aerosol during the same winter period (October, 1950, to January, 1951). For 21 plates exposed in the lower-floor 4bed ward the average counts for the three periods of the
exposed
na
gives
at intervals
TABLE IIA-AVERAGE
PLATES
EXPOSED
COUNTS DURING
OF BACTERIAL COLONIES FROM PERIOD FROM OCTOBER TO
A
JANUARY, 1951
TABLE IIB—AVERAGE COUNTS OF BACTERIAL COLONIES AFTER TREATMENT OF AIR WITH AEROSOL DURING A PERIOD FROM
OCTOBER TO
JANUARY, 1951