1. Background
About 60% of term and 80% of preterm infants have clinical jaundice in the first week after birth but only 0.02 to 0.16% of them develop severe hyperbilirubinemia (Total Serum Bilirubin (TSB) > 25 mg/dL), which is an emergency because it may cause neonatal bilirubin encephalopathy, which can result in death or irreversible brain damage in survivor (1-3). Intensive phototherapy and exchange transfusion (ET) play important roles in the treatment of severe hyperbilirubinemia of newborns to prevent bilirubin encephalopathy (4). Although the value of exchange transfusion in the treatment of neonatal hyperbilirubinemia is recognized, the bilirubin levels in which ET is necessary remained a matter of disagreement (5-7). ET is effective and considered to be a safe procedure; however, it is not without risks and complications have been reported and mortality rates vary from 0.5 to 3.3%. Therefore, the current recommendations for performing ET are based on balance between the risks of encephalopathy and adverse events related to the procedure (8, 9). Complications of ET may be increased by the amount of blood exchanged. Most of these complications are asymptomatic and transient, such as severe thrombocytopenia, apnea, hypocalcemia, bradycardia, and hyperkalemia, but life threatening infections and even death can occur within seven days after the exchange (10-13).
2. Objectives
The purpose of this study was to investigate the etiology and complications of exchange blood transfusion in a patient population visited in our center over the last five years.
3. Patients and Methods
The medical records of infants, 28 day old, who received exchange transfusion due to severe hyperbilirubinemia in neonatal units at Besat and Fatemieh university hospitals in Hamadan for a period of five years (from October 1st 2006 through September 30th, 2011 ) were reviewed retrospectively. Exchange transfusion was performed by pediatric residents under direct supervision of pediatric professor. The double volume exchange method (170 mg/kg) was completed for approximately 1 - 2 hours by repeatedly removing and replacing a small amount of blood (5 mL/kg) according to standard practice guidelines. Infants’ heart rate and oxygen saturation were monitored during exchange transfusion.
Laboratory investigations were performed such as complete blood counts, direct and total bilirubin, erythrocyte glucose -6- phosphate dehydrogenase (G6PD) level, direct coombs test, maternal and baby blood groups, serum calcium, glucose, sodium, potassium, and blood cultures before and after the exchange.
Intravenous calcium gluconate was used during exchange transfusion to neutralize the effect of citrate in Citrate Phosphate Dextrose Adenine (CPDA) solution. All episodes of complication were recorded up to three days after exchange transfusion. The definition of complications used were as hypoglycemia when serum glucose was < 50 mg/dL, hypocalcemia if serum calcium was < 8 mg/dL (for preterm newborn, 7 mg/dL), hyperkalemia when serum potassium was > 6 meq/dL, thrombocytopenia when platelet count was < 100.000/mm3, bradycardia if heart rate was < 80 beat/minute, apnea cessation of respiration for > 20 seconds, seizure any tonic or clonic movement, necrotizing enterocolitis defined as per bell's criteria. Complications of severe neonatal hyperbilirubinemia, consisting of kernicterus were not included in this study. Exclusion criteria were incomplete records of patients and those older than 28 days old. Statistical analysis was performed using SPSS Version 12.0 (SPSS, Inc, Chicago, USA).
4. Results
There were 6108 neonatal admissions due to neonatal jaundice that exchange transfusion was performed in 148 neonates (2.4%). Among them 68 (45.1%) were male and 80 (54.9%) female. Mean birth weight was 2847 ± 699 grams and mean gestational age of neonates was 37.2 ± 2.5 weeks. Overall, 102 (68.9%) infants were term and 46 (31.1%) preterm. The mean maximum total serum bilirubin was 27.76 ± 7.28 mg/dL and the mean age of exchange transfusion was 4.97 ± 2.65 days (Table 1).
Table 1. Baseline Demographic Characteristics a
| Characteristic | Value |
|---|---|
| Gender | |
| Male | 68 (45.1) |
| Female | 80 (54.9 ) |
| Gestational Age Group | |
| Term | 102 (68.9) |
| Preterm | 46 (31.1) |
| Gestational age, wk | 37.2 ± 2.5 |
| Maximum total serum bilirubin, mg/dL | 27.76 ± 7.28 |
| Age at exchange transfusion, d | 4. 97 ± 2.65 |
| Frequency of Exchange Transfusion | |
| One | 113 (76.4) |
| Two | 18 (12.2) |
| Three | 17 (11.4) |
| Complication rate | 57 (38.5) |
| Duration of hospital stay, d | 6.79 ± 6.64 |
a Data are presented as No. (%) or mean ± SD.
Among 148 cases, no etiologic factors were identified in 61 (41.2%) neonates, ABO incompatibility was found in 54 (36.5%), RH incompatibility in 15 (10.1%) and G6PD deficiency in 14 (9.5%) neonates (Table 2). During and immediately after exchange transfusion, 57 (38.5%) neonates developed complications. Most complications were thrombocytopenia (17.6%), hypocalcemia (11.5%), hypoglycemia (9.5%), hyperkalemia (5.4%), hyponatremia (4.1%), apnea (4.7%), and septicemia (2%). One (0.7%) neonate died of complications probably attributable to exchange transfusion (Table 3). Comparing variables between term and preterm infants showed no significant difference in complications between the two groups (Tables 4 and 5).
Table 2. Etiology of Neonatal Hyperbilirubinemia
| Causes | NO (%) |
|---|---|
| ABO incompatibility | 54 (36.5) |
| RH incompatibility | 15 (10.1) |
| ABO and RH incompatibility | 3 (2.0) |
| G6PD deficiency | 14 (9.5) |
| Polycythemia | 1 (0.7) |
| Unidentified | 61 (41.2) |
| Total | 148 (100) |
Table 3. Complications of Exchange Transfusion
| Complications | No (%) |
|---|---|
| Thrombocytopenia | 26 (17.6) |
| Hypocalcemia | 17 ( 11.5) |
| Hypoglycemia | 14 (9.5) |
| Hyperkalemia | 8 (5.4) |
| Hyponatremia | 6 (4.1) |
| Bradycardia | 12 (8.1) |
| Apnea | 7 (4.7) |
| Necrotizing enterocolitis | 3 (2) |
| Septicemia | 3 (2) |
| DIC | 2 (1.4) |
| Cardiorespiratory arrest | 2 (1.4) |
| Death | 1 (0.7) |
Table 4. Comparing Variables Between Term and Preterm Infants a
| Variables | Term 102 (68.9) | Preterm 46 (31.1) | P Value |
|---|---|---|---|
| Gender | 0.374 | ||
| Male | 44 (43.1) | 24 (52.2) | |
| Female | 58 (56.9) | 22 (47.8) | |
| Causes | 0.171 | ||
| ABO incompatibility | 42 (41.2) | 12 (26.1) | |
| RH incompatibility | 11 (10.8) | 4 (8.7) | |
| ABO&RH incompatibility | 2 (2) | 1 (2.2) | |
| G6PD deficiency | 11 (10.8) | 3 (6.5) | |
| Polycythemia | 0 (0.0) | 1 (2.2) | |
| Unidentified | 36 (35.6) | 25 (54.3) | |
| Complications | 0.529 | ||
| Yes | 39 (38.2) | 18 (39.1) | |
| No | 63 (61.8) | 28 (60.9) | |
| Thrombocytopenia | 0.400 | ||
| Yes | 19 (18.6) | 7 (15.2) | |
| No | 83 (81.4) | 39 (84.8) | |
| Hypocalcemia | 0.160 | ||
| Yes | 14 (13.7) | 3 (6.5) | |
| No | 88 (86.3) | 43 (93.5) | |
| Hypoglycemia | 0.238 | ||
| Yes | 8 (7.8) | 6 (13.0) | |
| No | 94 (92.2) | 40 (87.0) | |
| Hyperkalemia | 0.523 | ||
| Yes | 6 (5.9) | 2 (4.3) | |
| No | 96 (94.1) | 44 (95.7) | |
| Hyponatremia | 0.273 | ||
| Yes | 3 (2.9) | 3 (6.5) | |
| No | 99 (97.1) | 43 (93.5) | |
| Bradycardia | 0.545 | ||
| Yes | 8 (7.8) | 4 (8.7) | |
| No | 94 (92.2) | 42 (91.3) | |
| Apnea | 0.131 | ||
| Yes | 3 (2.9) | 4 (8.7) | |
| No | 99 (97.1) | 42 (91.3) | |
| Necrotizing enterocolitis | 0.232 | ||
| Yes | 1 (1.0) | 2 (4.3) | |
| No | 101 (99.0) | 44 (95.7) | |
| Septicemia | 0.324 | ||
| Yes | 3 (2.9) | 0 (0.0) | |
| No | 99 (97.1) | 46 (100.0) | |
| DIC | 0.526 | ||
| Yes | 1 (1.0) | 1 (2.2) | |
| No | 101 (99.0) | 45 (97.8) | |
| Cardiorespiratory arrest | 0.526 | ||
| Yes | 1 (1.0) | 1 (2.2) | |
| No | 101 (99.0) | 45 (97.8) | |
| Death | 0.689 | ||
| Yes | 1 (1.0) | 0 (0.0) | |
| No | 101 (99.0) | 46 (100.0) |
aValues are presented as No (%).
Table 5. Comparing Variables With and Without Complications
| Variables | Complications | P Value | |
|---|---|---|---|
| Yes = 57 (38.5) | No = 89 (61.5) | ||
| Gender a | 0.341 | ||
| Male | 29 (42.6) | 39 (57.4) | |
| Female | 28 (35.0) | 52 (65.0) | |
| Gestational age group a | 0.918 | ||
| Term | 39 (38.2) | 63 (61.8) | |
| Preterm | 18 (39.1) | 28 (60.9) | |
| Gestational age, wk b | 36.8 ± 3.0 | 37.5 ± 2.0 | 0.120 |
| Birth weight, g b | 2771 ± 764 | 2894 ± 656 | 0.299 |
| Admission age, d b | 4.6 ± 3.9 | 4.4 ± 2.3 | 0.844 |
| Maximum total Serum bilirubin, mg/dL b | 28.8 ± 8.6 | 27.1 ± 6.3 | 0.167 |
| Causes a | 0.145 | ||
| ABO incompatibility | 15 (27.8) | 39 (72.2) | |
| RH incompatibility | 8 (53.3) | 7 (46.7) | |
| ABO&RH incompatibility | 2 (66.7) | 1 (33.3) | |
| G6PD deficiency | 4 (28.6) | 10 (71.4) | |
| Polycythemia | 0 (0.0) | 1 (100.0) | |
| Unidentified | 28(45.9) | 33(54.1) | |
| Hospital stay, d b | 9.0 ± 7.2 | 5.3 ± 5.8 | 0.001 |
aValues are presented as No (%).
b Values are presented as mean ± SD.
5. Discussion
Exchange blood transfusion has reminded the gold standard for rapid lowering higher level serum bilirubin concentration and prevention of bilirubin encephalopathy and kernicterus. Although reports show progressive decline over the years in number of neonates who need exchange transfusion because of anti-Rh globulin for mothers and widespread use of phototherapy for neonatal jaundice, it is still required in up to 7% of neonates admitted to nurseries (13).
Despite advances in neonatal care in the recent years, exchange transfusion still remains a high risk procedure with common adverse effects. We observed a high rate of complications associated with exchange transfusion in 57 (38.5%) neonates; however, most of these were asymptomatic and transient. Most common complications in our study were thrombocytopenia (17.6%), hypocalcemia (11.5%), hypoglycemia (9.5%), hyperkalemia (5.4%), and hyponatremia (4.1%), which are similar to the findings of most previous studies (14, 15). Similarly, the rate of serious complication such as necrotizing enterocolitis and septicemia from ET is very low, approximately 1% and prior reports indicated that necrotizing enterocolitis and septicemia are the most common severe complications (11, 12, 16, 17).
Other serious complications of our study were apnea and bradycardia observed in 4.7% and 8.4% of neonates, respectively. Mortality directly attributable to exchange transfusion is reported to be at least 1% and is due to unexplained cardiac arrest, cardiac arrhythmias or air embolism (18). We observed a mortality of 1.4%; while, other studies reported a mortality rate range from 0.66% to 3.2% and (10, 12, 16, 19) Chime and Davutoglu reported no death in their study (20, 21).
Because many complications of exchange transfusion are unavoidable even with careful monitoring, early diagnosis of severe hyperbilirubinemia and phototherapy is the best way to reduce these complications that reduce the need for exchange transfusion in turn.
Multiple exchange transfusion was required in 23.6% of our neonates, which is similar to the findings of Dikshit (22), but more than Abu-Ekteish et al. (23). In our study, no etiologic factors were identified in 61 (41.2%) neonates, a rate reported previously as 17 - 40% (16, 24-27) and ABO incompatibility was observed in 54 (36.5%) neonates, which is similar to other studies (28-30). Rh incompatibility alone or concomitant with ABO incompatibility was observed in 15 (10.1%) and 3 (2.0%) neonates, respectively. The reduction in Rh incompatibility may be due to the use of anti-Rh globulin for Rh negative mothers (31). G6PD deficiency accounted for 14 (9.5%) of all causes of ET in our study. This figure is lesser than Badiee’s study (12), which estimated 19% prevalence of G6PD deficiency and higher than Bhat et al. (32) who reported no patient with G6PD deficiency. This difference in prevalence could be due to racial differences in the prevalence of G6PD deficiency.
This report indicated that complications are common after exchange transfusion despite technological advances in neonatal care and careful monitoring. Therefore, early recognition of infants at risk of severe hyperbilirubinemia and the use of intensive phototherapy can significantly reduce the need of exchange transfusion.
