Necrotizing Enterocolitis and Exchange Transfusion

Document Sample
Necrotizing Enterocolitis and Exchange Transfusion Powered By Docstoc
					 1236                                              S.A.     MEDICAL JOURNAL                                            21 July 1973

              Necrotizing Enterocolitis and Exchange
                           Transfusion *
  M.  SHAPIRO, D.SC., F.R.C.PATH. LOND., Director, South African Blood Transfusion Service, Johannesburg;
  H. STEIN, F.R.C.P. EDIN., D.C.H. R.C.P. LOND., R.C.S. ENC., Senior Paediatrician, Baragwanath Hospital, Johan-
  nesburg, AND A. OLINSKY, F.C.P. (S.A.), DIP. PAED. UNIV. R-LN"D, Senior Paediatrician, Queen Victoria and
                                 Transvaal Memorial Ho51J-itals, Johannesburg

                          SUMMARY                                          Idiopathic necrotizing enterocolitis unassociated with ex-
                                                                        change transfusion has also been recorded by many
    Most authors ascribe necrotizing enterocolitis following            authors'" as a not infrequent occurrence, particularly in
    exchange transfusion to a disturbance of the splanchnic             premature infants. With the recent increased awareness of
   circulation with superimposed infection. We wish to                  the clinical signs of the disease and especially of its dis-
   emphasize the primary role of intestinal infection, as               tinctive radiological features, it is being recognized earlier
   illustrated by outbreaks in 2 separate hospitals. Each               and more frequently than previously. A recent review of
   outbreak lasted about 6 weeks.                                       necropsy material by Hopkins et al." showed that idiopathic
                                                                        enterocolitis was present in 13 of 427 premature infants,
     Sixty-three babies in a prematurity unit of Baragwanath
                                                                        an incidence of 3,04%.
    Hospital had diarrhoea. Salmonella was cultured from
   the stools of- 22 of them. Eleven developed necrotizing                 Intestinal stasis and local ischaemia, at a time when
   enterocolitis and all of these died; 2 had received                  faecal organisms such as E. coli and Pseudomonas are be-
   exchange transfusions for hyperbilirubinaemia.                       ginning to colonize the bowel, have been suggested as likely
    The other outbreak occurred in the Queen Victoria
                                                                        causes. Concern has also been expressed regarding paren-
  Maternity Hospital. There were 19 cases of necrotizing
                                                                        teral feeding and medication via the umbilical vein as
  enterocolitis; 4 died. No specific pathogen was isolated.
                                                                        sources of infection.
  Thirteen of these babies had received exchange trans-
  fusions for hyperbilirubinaemia.
     In the 6 months embracing both outbreaks, 269 exchange                     SOUTH AFRICAN EXPERIENCE
  transfusions were performed by the same transfusion team
  in other units of Baragwanath Hospital, and 587 in                    In the years 1967 to 1969, approximately 4500 exchange
  maternity hospitals other than the Queen Victoria Hospital,           transfusions were performed by the South African Blood
  without a single case of necrotizing enterocolitis among              Transfusion Service in Johannesburg. .In this period, 5
  the 596 recipients.                                                   babies who had received exchange transfusions were ad-
     Although these outbreaks of necrotizing enterocolitis              mitted to surgical units with enterocolitis, an apparent
  were unquestionably due to intestinal infection ab initio,            incidence of this complication of 1: 900 exchange trans-
  the possibility remains that the exchange transfusions                fusions:· All 5 were mature neonates with hyperbilirubi-
  may have aggravated the necrotizing lesion in some                    naemia. One had Rh haemolytic disease.
  babies. It is therefore recommended that exchange trans-
                                                                           In the latter half of 1971, 2 outbreaks of necrotizing en-
  fusion be administered with caution, and preferably with-
                                                                        terocolitis occurred in Johannesburg -         one affecting
   held, if the infant has diarrhoea or blood in the stools,
                                                                        Bantu infants at the Baragwanath Hospital and the other
  or if intestinal infection is known to be prevalent in the
                                                                        affecting White infants at the Queen Victoria Hospital.
  unit or hospital.

  S. Afr. Med. J., 47, 1236 (1973).
                                                                        Baragwanath Hospital
Since 1968 there have been several reports of necrotizing                  In the case of the Baragwanath Hospital, I I the outbreak
enterocolitis following exchange transfusion in neonates,               was confined to one of the two prematurity units of the
mostly in cases of Rh haemolytic disease. ' ·' It appears to            hospital (Table I). Sixty-three babies in this unit had
have been accepted by these authors that the bowel                      diarrhoea. In 22 of them Salmonella organisms were cul-
necrosis in such cases was due to interference with the                 tured from the stools. Necrotizing enterocolitis was diag-
splanchnic circulation by the transfusion procedure, and                nosed in 11 cases. All the infants affected with necrotizing
that erythroblastotic infants are especially susceptible to             enterocolitis died. The intervals between the onset of the
this complication.                                                      diarrhoea and signs of enterocolitis varied from 1 to 21
                                                                        days. Five of the 11 affected infants had significant jaun-
'Paper presented at the 13th Congress of the International Society of
 Blood Transfusion, Washington. DC, USA, September 1972.                dice with bilirubin levels above 10 mg/l00 m\. In 2 of
21    Julie    1973                               S. - A.     MED lE SE TYDSKRIF                                                          1237

      Jaundice                                                                       Onset of
      (SB level                                           Onset of                  necrotizing           Exchange             Interval from
        >10                      Salmonella               diarrhoea                enterocolitis         transfusion           last exchange
     mg/l00 ml)                    isolated             (age in days)             (age in days)           (number)                  (days)
         Yes                         No                      23                        29                    No

         No                          Yes                     11                        32                    No
         Yes                         Yes                      3                         6                    No

         No                          No                       3                         4                    No
         Yes                         No                       6                        12                    No

          No                         Yes                      9                        19                   No
          No                         Yes                     10                        18                   No
          No                         Yes                     10                        11                   No
         Yes                         Yes                      3                         5                 Yes (1)                    2
         Yes                         No                       3                         6                 Yes (2)                    2
          No                         No                      11                        25                   No

them the serum bilirubin levels exceeded 17 mg/lOO ml,                       demonstrated In one of these two infants. In each case
and both received exchange transfusion - one once, and                       blood was observed in the stools 2 days after the last
the other on two occasions. Salmonella organisms were                        exchange transfusion. It seems unquestionable that in both



                                  Peak        Mother-                                         Interval                   Outcome
               Birth             serum         infant      Exchange                          from last
              weight            bilirubin       ABO          trans-                    Age exchange                    Recovered
 Sex           (kg)     Apgar (mg/l00ml)      groups        fusions      Date         (days) (days)      Operation      or died     Remarks
  M            2,7        9        17,1        O/B                2     22 Nov.         8          5       Yes         Died        E. coli
  M            1,9        8        19,2       B/AB                3     28   Nov.      16      10          Yes         Recovered
  M            2,0        8        18,5       A/O                 1      3   Dec.       5       1          Yes         Recovered
  M            3,4        8        18,5       B/B                 1      4   Dec.       6       2          No          Recovered
  M            1,6        8        17,5       O/B                 1      6   Dec.      18      14          No          Died        E. coli
  F            1,4        3        19,0        0/0                2      7 Dec.         8          2       No          Died        Gangrene
                                                                                                                                   of colon
  F            2,7        6        21,1       O/B                 2     10   Dec.       9          7       No          Recovered
  F            4,2        9        12,8       AB/A                0     10   Dec.       7                  No          Recovered
  F            1,9       10        19,7       A/A                 1     11   Dec.       6                  No          Recovered
  M            2,1        8        19,0       A/A                 1     11   Dec.       5                  No          Recovered
  M            3,2        9        21,2       O/B                 1     11   Dec.       6                  No          Recovered
  F            1,9        6        Not        ?/O                 0     11   Dec.       2                  No          Recovered
 M             1,9       10        18,1        AB/A               1     13   Dec.       7          3       No          Recovered
 F             2,3        6        20,5        A/A                1     15   Dec.       5          6 h     No          Recovered
 F             3,1        8        11,3        A/O                0     16   Dec.       3                  No          Recovered
 M             3,1        2         8,8        A/A                0     16   Dec.       3                  No          Recovered
 M             2,5        9        17,2        O/A                0     16   Dec.       8                  No          Recovered
 M             1,7        4        20,0        ?/O                1     28   Dec.       6          2       No          Died        E. coli
                                                                                                                                   with brain
 M             2,2        8         9,9        A/A                0     16 Feb.         2                  No          Recovered
1238                                        S. A.    MED J CA L          J   0 URN A L                                       21 July 1973


                                                  1 Oct. - 15 Nov.                    16 Nov. - 31 Dec.                 1 Jan. - 15 Feb.

                                                Idiopathic     Rh HDN                 Idiopathic     Rh HDN          Idiopathic        Rh HDN
                                                    10           4
No enterocolitis   {   Transfused
                       Not transfused                6           0
                       Transfused                    0           0                       14            0                     0           0
With enterocolitis {
                       Not transfused                0           0                        0            0                     0           0

                                                    16           4                       37            5               15                2
                                                          20                                   42                                17

                           Total births                  552                                  531                                523

of these cases infection was the primary event, and not                                             DISCUSSION
[he exchange transfusions.
                                                                     In the 6 .months embr~cing both outbreaks of necrotizing
Queen Victoria Hospital                                              enterocolitiS reported In this paper, 269 exchange trans-
                                                                     fUSIOns were performed by the same transfusion team in
                                                                     other units of the Baragwanath Hospital and 587 in
   The outbreak at the Queen Victoria Hospital occurred
                                                                     maternity hospitals other than the Queen Victoria Hos-
about 2 months later (Table 11). During the 6 weeks'
                                                                     pital. There were no cases of necrotizing enterocolitis
period 18 infants were affected with necrotizin cr entero-
                                                                     among the 596 recipients of these transfusions. There can
coli[is, 9 of them premature babies. One additi;'nal case.
                                                                     be no doubt that infection was the primary cause of the
also a premature, occurred about 5 weeks later and may
                                                                     2 outbreaks of necrotizing enteroco;itis reported here the
have been unrelated to the outbreak. Six of the affected
                                                                     predominant ear;y signs being diarrhoea in all the B~racr­
infants received no exchange transfusions. Fourteen
                                                                     wanath Hospital cases and hyperbilirubinaemia in the
developed serum bilirubin levels in excess of 17 mg/loo
                                                                     majority of cases at the Queen Victoria Hospital. The
ml and all but I of these received exchange transfusions.
                                                                     dIrect correlation of exchange transfusion is with the
  lone of the affected infants suffered from Rh haemolytic
                                                                     hyperbilirubinaemia which necessitated the procedure and
diseas::. In those infants who received transfusions the
                                                                     nor wif, t',e necrotizing enterocolitis with which the
intervals between the last exchange and the f"irst ap~ear­
                                                                     hyperbilirubinaemia was associated in some of the cases
ance of necrotizing enterocolitis varied from 6 hours to
                                                                     Fifteen of the 30 infants with necrotizing enterocolitis i~
 14 days. Three of the transfused infants were operated on
                                                                     these 2 outbreaks received no exchange transfusions.
for bowel perforation. Of these, I died. Altogether 4 of
                                                                       Tevertheless, in considering the pathocrenesis of the disease
                                                                     it is ~onceivab:e that a disturbance ;f the splanchnic cir~
the 19 affected infants died. At necropsy, E. coli meningitis
was found in addition to the bowel lesion in 3 of these
                                                                     culatlon, associated with an exchange transfusion, may
infants. One had multiple brain abscesses as well. Thr. e            aggr::: vate a nocrotlzmg process in the bowel which is
of the 4 who died were premature infants.                            already devitalized by inflammation or ischaemia. Con-
   No specific pathogen was recovered from the stools of             siderations of prudence would therefore appear to dictate
,:ny of the affected babies and the infection in this in-            [hat exchange transfusion should be administered with
stance may have been viral in· origin.                               caution, and preferably withheld, if the jaundiced infant
    A feature of this epidemic was the increased incidence           has bl~od in the stools, persistent or bile-stained vomiting,
of hyperbilirubinaemia at this hospital in the 6 weeks'              abdomInal dlstenslOn or suspicious radiological signs of
period during which these cases occurred (Table Ill). In             necrotIZIng enterocolitis, or if diarrhoea is known to be
this period, 37 neonates in the Queen Victoria Hospital              prevalent in the particular unit or hospital.
developed serum bilirubin levels in excess of 17 mg/IOO
ml, and of these, 14 were affected by entero:olitis. By                                              REFERE 'CES
contrast, there were only 16 cases of hyperbilirubinaemia            I. COSIer, w. R. (1969): Canad. Med. Assoc. J., 99, 934
in this hospital in the 6 weeks preceding the epidemic and 15        2. Corkery, J. J., Dubowi!z, V., lisler, J. and Moosa A (1968): Brit.
                                                                        Med. J., 4, 345..                                  ' ..
cases in the corresponding period following the epidemic.            3. Orme, L'E. and Eades, S. M. (1968): Ibid., 4, 350.
                                                                     4. j;;~ldhausen, J. A .. Herendeen, J. and King, H. (1963): Surgery, 54.
The difference is statistically highly significant (x' = 13.3;
P<0,0005). The prev3.lenc; of hyperbilirubinaorr.ia durirg            5. ~~~?~\~: a~~ BT~~c,G~ss,;:an(19~4):B;:~r~IO~Y, ~j: lWzrahi, A. Olive
r:,e period w:,en the outbreak of necrotizing enterocolitis           6. Mlzrahl, A., OlIve Barlow, Berdon. \V., Blanc. \V. A and Silverman
                                                                         W. A. (1965): J. Pediat., 66, 697.                    .             ,
occurred, suggests a fairly widespread viral infection with           7. Brennan M. F. (1967): N.Z. Med. J., 66. 385.
hyperbilirubinaemia as an early sign, in many cases with              8. Slevenson, J. K., Graham, C. B., Oliver, T. K. and Goldenbere.
                                                                         V. E. (1969): Amer. J. Surg., 118. 260.                           -
necrotizing lesions of the bowel as a later manifestation. It         9. Hopkins, G. B., Gould, v. E.. Slevenson J. K. and Oliver T K
                                                                         (I970): Amer. J. Dis. Child., 120. 119. .                    , .    .
is perhaps worth noting that none of the 5 infants with              10. ~~ck. J. M., Dinner, M. and Chappel, J. (1971): S. Afr. J. Surg., 9,
erythroblastosis born during the outbreak and requiring              11. Stein, H., Beck, J., ,Solomon. A. and Schmaman. A. (1972):          Brit.
exchange transfusions, developed enterocolitis.                         Med. J., 2, 616.