- Email: [email protected]
Cold Injury Syndrome and Neurodevelopmental Changes in Survivors
Archives of Medical Research 36 (2005) 532–538
ORIGINAL ARTICLE
Cold Injury Syndrome and Neurodevelopmental Changes in Survivors Srdjana Culic Department of Pediatrics, Clinical Hospital Split, Split, Spinciceva, Croatia Received for publication July 23, 2004; accepted March 18, 2005 (ARCMED-D-04-00041).
Background. The aim of this study was to extend and develop knowledge of the clinical parameters of cold injury syndrome (CIS) in hypothermic infants and newborns, so that morbidity, mortality, and neurodevelopmental disturbances associated with the condition can be reduced. Methods. This retrospective, 10-year cohort study investigated 103 hypothermic infant and newborn patients (aged 0–60 days) admitted to the Pediatric Department between January 1, 1976 and December 31, 1985. Medical records of 67 survivors and 36 children who died were analyzed. Of the survivors, 24 consented to undergo assessment of neuropsychological and neurodevelopmental outcomes. Results. At presentation, all patients manifested cold skin and limbs, while 96 (93.2%) were sleepy, hypotonic and inactive. Cyanosis with bradycardia, bradypnea, and apneic crisis (AC) was observed in 80 (77.7%) patients, and 65 (63.1%) demonstrated diffuse intravascular coagulation (DIC). Pulmonary hemorrhage, AC, respiratory distress syndrome, and DIC represented the primary causes of death. Higher body temperatures on admission reduced the risk of death. Of the 24 patients assessed for neuropsychological and neurodevelopmental outcomes, all but one had normal intellectual capabilities, whereas 16 (66.7%) showed signs of neurodevelopmental disturbances. Conclusions. The severity of hypothermia, leukopenia, and thrombocytopenia with DIC correlates with prognosis and death rate. Hypothermia in infants and newborns can cause psychological and neurodevelopmental disturbances in survivors. 쑖 2005 IMSS. Published by Elsevier Inc. Key Words: Hypothermia, Infants, Diffuse intravascular coagulation, Neurodevelopment.
Introduction Hypothermia is defined as a core body temperature (BT) of 35⬚C or lower. It can be classified according to one of the following categories: mild (BT: 32.2–35⬚C); moderate (BT: 28–32.2⬚C); and severe (BT: ⬍28⬚C). Due to the inadequacy of standardized thermometers, hypothermia frequently remains unrecognized as its accurate diagnosis requires a low-reading thermometer. The hypothermia syndrome is frequently life-threatening and is characterized by multiple Address reprint requests to: Srdjana Culic, PhD, Clinical Hospital Split, Pediatrics Clinic, 21000 Split, Spinciceva 1, Croatia; E-mail: srdjana. [email protected]
0188-4409/05 $–see front matter. Copyright d o i : 10 .1 0 1 6/ j.ar c med .2 0 05 .0 3 .0 4 3
causative factors and a diverse spectrum of pathophysiological consequences. Cold injury syndrome (CIS) may be described as a type of accidental hypothermia recognized in our region (Croatia) and in Israel (1,2). The term cold injury syndrome was first used by Mann at the Royal Society of Medicine meeting in 1953 to describe the condition observed in pre-term, newborn infants exposed to cold (2–4). Between 1955 and 1957, Mann and Elliot (4) and Smallpiece and Jones (5) published a great deal of new information on CIS. From reviews of this and other reports in the literature, it is evident that CIS is a unique clinical entity occurring in infants and neonates with hypothermia.
쑖 2005 IMSS. Published by Elsevier Inc.
Cold Injury Syndrome
Interview protocols for the assessment of psychological and neurodevelopmental outcomes in the 24 survivors who consented to further assessment included evaluation of both intellectual capabilities and signs of neurological dysfunction. For children ⬎4 years, the Kohs test (9) and the Binet⫺ Simone scale (10) were used and visual–motor perception was assessed using the Bender–Gestalt test (11). In order to evaluate psychological development in children ⬍4 years of age, the Brunet–Lezine scale (11) was applied. In addition, a psychological interview was carried out by the hospital psychologist. Following psychological and neurodevelopmental evaluation, the 24 survivors were categorized into one of three groups: 1) patients with normal psychological functioning; 2) patients with normal intellectual development and capabilities, but showing signs of neurological disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual perception, and other specific cognitive deficits and developmental retardation, e.g., enuresis; 3) patients with lower intellectual capabilities.
Results From January 1, 1976 to December 31, 1985, the overall number of hospitalized children was 17,753, with a total of 266 (1.5%) deaths. CIS was diagnosed in 103 children, and 36 (34.95%) of these patients died. We calculate that the frequency of this disease during the period between January 1976 and December 1985 was 0.58% with a mortality rate of 13.53%, which represents a significant proportion of patients. Children suffering from CIS were most often newborn babies; 80 (77.67%) patients were between 0 and 15 days old. Figure 1 shows the ages at admittance of those patients
50 45
Children suffering from CIS
40 35
Children dying from CIS
30 25 20 15 10 5
-5
5
5 41
-4
5 31
-3
5 -2 21
11
-1
5
0 5
The databases of the Pediatric Department of the Clinical Hospital in Split were searched for patients diagnosed with CIS between January 1, 1976 and December 31, 1985. During this period, 103 children with signs of both hypothermia and subsequent organ injuries presented and were treated at the Pediatric Department. None of the CIS patients had coexisting or predisposing conditions. The clinical and laboratory records of these patients were retrospectively reviewed. Additional data were collected from 24 CIS survivors who responded, or whose parents responded, to an initial telephone poll and consented to undergo psychological and neurodevelopmental evaluation.
0-
Subjects and Methods
51
Severe hypothermia causes multiple systemic disturbances that in turn may lead to decreased tissue oxygenation, circulatory collapse, hepatic and renal failure, dehydration, and inappropriate activation of the coagulation cascade; the latter may precipitate diffuse intravascular coagulation (DIC) (6). Decreased oxygen consumption and metabolic rate may precede hypothermia-induced endothelial damage and subsequent consumption coagulopathy. Investigating the effects of temperature change on the coagulation time of blood, Kmiecik et al. (7) reported that activated clotting time increases as body temperature decreases from 37⬚ to 27⬚C. Severe hypothermia produces cardiac complications, apnea, fixed and dilated pupils, and an isoelectric EEG, and attempts to resuscitate the patient by increasing body temperature often fail (8). Infants with CIS cool slowly and are usually hospitalized before such complications occur. Where decrease in body temperature occurs rapidly to ⬍30⬚C, organ ischemia may be avoided. The current study supports the hypothesis that CIS is a distinct clinical entity, showing patterns of national, and particularly regional, pathology. Analyses performed during this study are based on documentation from infants treated for CIS at the Pediatric Department of the Clinical Hospital in Split between 1976 and 1985. Data were collected from patient records and from telephone polls completed by those CIS survivors who agreed to provide information. The aim of the study was to evaluate 1) all patients with CIS who were treated in the Pediatric Department at the Clinical Hospital in Split during the 10-year period from January 1, 1976 to December 31, 1985; 2) the exogenous factors that trigger the onset of the syndrome; and 3) the neurodevelopmental outcome of CIS survivors. We hypothesized that hypothermia may induce neurodevelopmental disturbances, including one or more of the following symptoms: deficits in attention, concentration, visual–motor perception; hyperactivities; disturbances in motor coordination; altered cognitive functioning; and developmental retardation. In addition, we sought to analyze the pattern of hypothermia, its connection with DIC, and its effects on organ systems in neonates and infants.
533
Figure 1. Age (days) at admittance of patients with CIS.
534
Culic / Archives of Medical Research 36 (2005) 532–538
who suffered and/or died from the condition. There were 61 (59.22%) boys and 42 (40.78%) girls; 47 (45.63%) of these babies were full-term, whereas 45 (43.69%) were pre-term. In addition, 86 (83.49%) patients had a birth weight of ⬎2500 g. A total of 78 (75.73%) patients came from the village regions; of these, 28 (27.18%) and 22 (21.36%) patients were from the surrounding regions of the towns of Sinj and Split, respectively, whereas 12 (11.65%) patients came from the town of Split itself. Children were hospitalized more frequently in the winter than in the summer months, with 32 (31.07%) and 21 (20.39%) patients admitted in December and January, respectively. During the summer, there were no reported cases of CIS. In addition, the number of children suffering and/or dying from CIS decreased during the period between 1976 and 1985 (Figure 2). Frequently, CIS patients came to the department in serious condition, with signs of hypothermia and organ injuries already evident. A severe medical condition on admittance was observed in 84 (81.6%) infants, whereas 19 (18.5%) patients were considered to be in moderate condition. All children manifested cold skin and limbs along with red facial skin, whereas 96 (93.2%) were sleepy, hypotonic and inactive. Furthermore, 80 (77.66%) patients were cyanotic with bradycardia and bradypnea, while 77 (74.76%) manifested scleredema. Other symptoms included jaundice, apneic crisis (AC), vomiting, seizures, and secondary infections (Table 1). The lowest rectal body temperature recorded at admission was 21⬚C, although rectal temperatures ranged from 21⬚ to 36⬚C. It was observed that the higher the rectal temperature at admission, the lower the risk of death (Figure 3). According to the anamnestic data, the duration of the condition was very short, often lasting for only 1 day and 23 (63.8%) of the 36 deaths occurrred within the first 24 h
18
Table 1. Symptoms shown by patients with CIS admitted between January 1, 1976 and December 31, 1985 Symptom
Number of patients (%)
Cold skin and extremities Red facial skin Adynamia Lethargy Hypotonia Anorexia Cyanosis Scleredema Bradycardia Bradypnea Jaundice Apnea Vomiting Infection Seizures
103 103 96 96 96 95 80 77 70 69 61 38 30 29 21
(100) (100) (93.2) (93.2) (93.2) (92.2) (77.7) (74.7) (67.9) (66.9) (59.2) (36.8) (29.1) (28.1) (20.3)
(Table 2). Additionally, 20/47 (42.55%) full-term babies died compared with 14/45 (31.12%) pre-term babies (Table 3). Of the 17 patients with a birth weight of ⬍2500 g, 10 (58.82%) died, whereas of 86 patients with a birthweight of ⬎2500 g, 26 (30.23%) died (Table 4). Glycemia was detected in 66 (64.07%) patients, whereas 42 (40.77%) children were found to be hypoglycemic. In 65 (63.10%) CIS patients, DIC was observed, 21 (26.58%) patients had thrombocytopenia with DIC, and 36 (45.57%) demonstrated both thrombocytopenia and leukopenia with DIC (Table 5). In addition, 87 (84.47%) children were polyglobulic. Due to technical reasons, antithrombin III (AT III) levels were analyzed in only four patients at admission; these four patients showed AT III levels of 60%, 34%, 35%, and 24%, respectively. Most (28/36; 72.22%) patients who died as a result of CIS were ⬍15 days old. All children who died had metabolic acidosis and AC due to pulmonary hemorrhage. There were
12
16 10
14 12 10 8
Children suffering from CIS
8
Children dying from CIS
6
Children suffered from CIS Children died from CIS
4
6 4
2
2 0
0 1976
1978
1980
1982
1984
Figure 2. The number of children suffering/dying from CIS between January 1, 1976 and December 31, 1985.
2122
2324
2526
2728
2930
3132
3334
3536
Figure 3. Body rectal temperature (⬚C) at admittance of children suffering/ dying from CIS.
Cold Injury Syndrome
535
Table 2. Length of time to death in CIS patients following admittance
Table 4. Incidence of CIS and death: birth weight
Length of time to death (h)
Birth weight (g)
Number of patients (%)
24 48 72 ⬎72 Total
23 7 1 5 36
(63.8) (19.4) (2.7) (13.8) (100)
no differences in death rates between boys and girls. Seven (87.5%) of eight patients with a temperature ⬍25⬚C died, whereas 50% of children with a temperature between 25.1⬚ and 26⬚C survived (Figure 3). Further analysis of deceased patients revealed that 29 (80.55%) children had a low white blood cell (WBC) count, with 19 (52.78%) children demonstrating a WBC level of ⬍2.0 × 109/L (normal levels are 4–10 × 109/L). Neurodevelopmental assessment in the 24 CIS survivors revealed that 7 (29.17%) patients had had normal neuromotor development, whereas 16 (66.67%) showed normal intellectual capabilities with signs of neurodevelopmental disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual perception, as well as other specific cognitive deficits and developmental retardation (Figure 4). Only one (4.16%) patient, born as a pre-term baby, demonstrated diminished intellectual capacity. Further analyses found no correlation between gestational age and occurrence of neurodevelopmental disturbances. Only one (4.17%) post-term child with normal intellectual capabilities and neurodevelopmental disturbances was observed (Figure 4).
Discussion Dragovich (12) has described how hypothermia is a common problem in neonates and an important contributory factor in neonatal morbidity and mortality, particularly in developing countries where knowledge and practice of the thermal control of newborns is insufficient. Increased awareness and improved knowledge of thermal regulation and thermal protection in such countries may help to prevent CIS in both newborn babies and young infants, whereas improved understanding of its clinical manifestations may hasten diagnosis and aid effective treatment. We report here on the demographic parameters and clinical and laboratory presentations of CIS in neonates and Table 3. Incidence of CIS and death: gestational age
Gestational age
Pre-term (%)
Full-term (%)
Post-term (%)
Total (%)
Incidence of CIS Incidence of deaths
45 (43.6) 14 (38.8)
47 (45.6) 20 (55.5)
11 (10.6) 2 (5.5)
103 (100) 36 (34.9)
1000–1500 1501–2000 2001–2500 2501–3000 3001–3500 3501–4000 4001–4500 Total
Number of patients (%)
Number of deaths (%)
1 (0.9) 7 (6.8) 9 (8.7) 45 (43.7) 32 (31.1) 8 (7.8) 1 (0.9) 103
1 (2.8) 5 (13.9) 4 (11.1) 13 (36.1) 9 (25) 4 (11.1) 0 36
infants, as well as potential neurodevelopmental consequences in survivors of this condition. Our study revealed that the death rate of CIS patients was highest in December and November, which are cold months during which patients tended to be particularly severely affected. At the time of admittance the lowest rectal temperature recorded was 21⬚C, and rectal temperatures ranged from 21⬚ to 36⬚C. Higher body temperatures at admittance were associated with a reduced mortality rate; however, one infant with a rectal temperature of 22⬚C did survive. Rate of death was independent of birth weight or gestational age (Table 4). The cause of death in patients with CIS is almost certainly thrombocytopenia with DIC with or without leukopenia, which results from hypothermic stress and shock. Normal hemostasis is a dynamic balance between bleeding and thrombosis (13), and DIC is considered to be a systemic failure of coagulation, leading to consumption coagulopathy and a tendency to bleeding and multiple thrombi causing dysfunction of one or multiple organs. In addition, platelet contact induces a variety of leukocyte reactions and results in the production and secretion of several cytokines and adhesive proteins associated with the inflammatory response. Activated coagulation factors and cytokines such as interleukin 1 (IL-1) or tumor necrosis factor (TNF) also mediate
Table 5. Incidence of death: WBC, PLT, and DIC Number of patients (%) Thrombocytopenia and DIC Thrombocytopenia without DIC Leukopenia and DIC Leukopenia without DIC Thrombocytopenia, leukopenia without DIC Thrombocytopenia, leukopenia, DIC Decrease in thrombocyte numbers without thrombocytopenia, DIC Total
21 3 1 4 7
Number of deaths (%)
(36.70) (3.79) (1.26) (5.26) (9.21)
7 (33.3) 0 0 1 (25) 3 (42.8)
36 (45.56)
22 (61.1)
7 (8.86) 79
0 33 (41.7)
WBC, white blood cells; PLT, platelet numbers; DIC, diffuse intravascular coagulation.
536 16 14 12 10 8 6 4 2
Culic / Archives of Medical Research 36 (2005) 532–538 Normal psychological functioning Normal intellectual capabilities with signs of neurodevelopmental dysfunctions Low intellectual capabilities
Table 6. Incidence of death: platelet levels PLT 109/L
Number of patients (%)
⬎110 110–70 69–30 ⬍30 Total
27 30 29 17 103
Number of deaths (%)
(26.21) (29.12) (28.15) (16.50)
3 12 13 8 36
(11.1) (40.0) (44.8) (47.7) (34.9)
PLT, platelets.
0
Figure 4. Neurodevelopmental outcomes of CIS survivors.
activation of endothelial cells and lead to intravascular fibrin deposition (14). Russwurm et al. (15) have investigated the influence of hypothermia on cytokine expression and concluded that secretion of IL-2, IL-6, IL-10, and TNF-α was reduced in hypothermic cultures of peripheral blood mononuclear cells. The authors suggested that reduction of IL-2 production might impair the immune response leading to an increase in infectious complications. The pathophysiological manifestation of DIC is a dynamic process that occurs in three phases. While in the first phase hemostasis is compensated, in the second phase uncompensated activation of the hemostatic system occurs and is associated with decreased organ function and bleeding from injuries and venous puncture sites. DIC becomes fully expressed during the third phase, when it is characterized by skin bleeds, multi-organ failure, and an extremely prolonged prothrombin time (PT) and activated partial thromboplastin time (aPTT). In addition, secondary fibrinolysis, which inhibits vascular occlusion, induces consumption of platelets, leukocytes, coagulation factors, and inhibitors. During this phase, platelet and leukocyte counts are very low, and coagulation factor activities are ⬍50% lower than normal activity levels. These observations are supported by data from animal studies. Ao and co-workers (16) concluded that, in dogs, long-term hypothermia induces platelet dysfunction leading to decreased platelet aggregation and prolonged coagulation time. We revealed platelet dysfunction by a platelet function analysis with standard platelet aggregation tests that we performed in just one patient. Based on the aggregograms examined by standard Born method (17), in this patient platelet function analysis by ADP as induction substance revealed the block of endogenous ADP release. When we induced platelet aggregation via another stimulus such as adrenalin, we found that aggregation was absent. Technical limitations prohibited further analysis of platelet function in our patients. In the present study, the majority of children (81.5%) presented in serious condition, with signs of hypothermia already expressed and organ injuries probably in the second or third phase of DIC. Analysis of laboratory tests revealed a high frequency of leukopenia and thrombocytopenia, which were generally associated with DIC. Tables 6 and 7 show
the correlation between platelet and white blood cell counts and the incidence of death; these findings suggest that the leukocytes in the microvascular thrombi were consumed. In addition, we noted that in a small percentage of cases selective bone marrow injury caused by hypothermia and producing thrombocytopenia (3 [3.79%] patients) or leukopenia (4 [5.26%] patients) without DIC. Criteria for diagnosis of DIC have yet to be established. Wada et al. (18) observed hemostatic abnormalities prior to the onset of DIC and suggested that the monitoring of the plasma FDP-D-dimer in pre-DIC conditions may be beneficial. They further suggest that the prognosis of patients with DIC may be related to organ failure or endothelial cell damage (19). Imbalances in the cytokine network and its influence on coagulation have also been observed (20). Future investigations of cytokines such as IL-6, TNF-α, and IL-1β in children with CIS may help to explain the underlying pathophysiology of the syndrome. In any case, the degree of leukopenia and thrombocytopenia with the presence of DIC forms a laboratory parameter important for the estimation of prognosis and death rate. AT III is a physiological inhibitor of activated clotting factors. If the available biological activity of AT III diminishes below approximately 50% of normal activity, the inactivation of clotting enzymes is greatly inhibited. In three of four patients with DIC in the current study, AT III levels were significantly diminished. In such cases, it is important to substitute the depleted AT III (21). In the majority of survivors, neurodevelopmental outcome was assessed at 3–10 years of age. We observed that 16 of 24 (66.67 %) patients showed signs of neurodevelopmental disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual
Table 7. Incidence of death: white blood cell levels WBC 109/L
Number of patients (%)
⬎5.0 5.0–3.5 3.4–2.0 ⬍2.0 Total WBC, white blood cells.
38 27 17 21 103
(36.89) (26.21) (16.50) (20.38)
Number of deaths (%) 7 3 7 19 36
(18.4) (11.1) (41.1) (90.4) (34.95)
Cold Injury Syndrome
perception, and other specific cognitive deficits and developmental retardation, although intellectual capabilities in these patients were not impaired. Recent studies (22,23) investigated the influence of hypothermic circulatory arrest as a means of supporting vital organs during open-heart surgery in infants. These reports highlight the presence of a strong association between neurodevelopmental abnormalities and the duration of deep hypothermia, although this association was not evident in all cases. Hall et al. (24) investigated cold-induced platelet aggregation in platelet-rich plasma and concluded that platelet aggregates in the blood of donors with a high number of long-lasting aggregates were rigid and occlusive. The authors suggested that such aggregates may contribute to the cognitive dysfunction noted in patients undergoing hypothermic open-heart surgery. These studies indicate, as does our own experience, that the brain is extremely vulnerable to deep hypothermia in early infancy. Optimal treatment of CIS patients was not assessed in this study. However, we suggest that the optimal treatment for CIS is comprised of the re-warming of the patient to combat the underlying condition and the administration of high doses of corticosteroids. Such a regimen targets both RDS and DIC which, together, represent the primary causes of death in CIS patients. As soon as possible after presentation, a deep body (rectal) temperature reading should be obtained and the patient should be placed under a preheated radiant heater or inside a closed incubator. Active external rewarming is usually sufficient if the core temperature is ⬎32⬚C. However, active core rewarming should also be performed in patients with a core temperature of ⱕ32⬚C. The current trend in active external re-warming is to rewarm slowly. Our recommendation is to re-warm at a rate of 1⬚C/h, starting at 29⬚C, using a radiant heater or a closed incubator. Core re-warming involves the use of warmed (37⬚C) intravenous infusions and extracorporeal blood warming. The patient’s vital signs should be frequently monitored by personnel trained in pediatric intensive care. Whether the rate of re-warming is related to survival rate in hypothermic infants remains controversial, and there are no data in the literature pertaining to the effects of re-warming rate on CIS. An experimental model of DIC used to assess the preventive effects of methylprednisolone suggests that this drug may be beneficial in the treatment of conditions that induce DIC, such as CIS (25). Yamazaki et al. (26) have shown that prednisolone is more effective than heparin in reducing mortality in DIC patients. Our recommendation for the treatment of hypothermic infants is comprised of the use of high doses of methylprednisolone, low-molecular-weight heparin and antithrombin III, as well as re-warming. One of the more recently available treatment options for DIC is recombinant acivated factor VII (rFVIIa, NovoSeven), which has shown efficacy in controlling the acute bleeding in adults and children that results from the consumption of clotting factors in the later phases of DIC (27,28).
537
The frequency of reports of CIS in the hospital database reflects behavioral customs and insufficient home heating to which many infants at this time were subject. Even now, despite warmer homes and changes in midwifery practice, infants are still being admitted for hypothermia in our region. For example, in the period between January 1986 and December 1995, 29 infants with CIS were treated in our clinic. Of these, nine (31%) children died. In contrast, between 1991 and 1995, only five infants were hospitalized with CIS and all survived. This suggests that, although cases of CIS still occur, the incidence of the syndrome is decreasing. Some authors (29) recommend that the Neonatal Hypothermia Indicator should be used in order to prevent hypothermia and CIS in infants. Additional strategies for preventing CIS include parent education and sufficient home heating. Early diagnosis, appropriate re-warming and urgent administration of therapy may prevent psychological and neurodevelopmental disturbances associated with CIS. As Goldsmith and colleagues (30) suggest, excess neonatal mortality during winter months, particularly those deaths attributed to pneumonia or sudden infant death syndrome (SIDS), may be an indicator of missed CIS. In conclusion, the degree of leukopenia and thrombocytopenia in CIS, together with the presence of DIC, forms an important laboratory parameter for the estimation of prognosis and death rate. Early diagnosis is essential for the treatment of DIC. We conclude that intellectual development in children with CIS is generally within the normal range, although neurodevelopmental abnormalities may occur as a result of cerebral hypothermia.
References 1. Stambuk R. Clinical, pathoanatomical features and therapy for cold injury syndrome in children. Ph.D. thesis, Zagreb, Croatia, 1974. 2. Cohen IJ, Amir J, Gedaliah A, Rachmal A, Gorodischer R, Zaizov R. Thrombocytopenia of neonatal cold injury. J Pediatr 1984;104: 620–622. 3. Mann T. Hypothermia in the newborn: a new syndrome. Lancet 1955;2:613–614. 4. Mann T, Elliot RIK. Neonatal cold injury due to accidental exposure to cold. Lancet 1957;1:229–234. 5. Smallpiece W, Jones E. Hypothermia in the newborn. Lancet 1955;1:22. 6. Lazar HL. Treatment of hypothermia. N Engl J Med 1997;337: 1545–1547. 7. Kmiecik SA, Liu JL, Vaadia TS, Nichols JD, Kohtz RJ, Mills NJ, Petterson CM, Stammers AH. Quantitative evaluation of hypothermia, hyperthermia, and hemodilution on coagulation. J Extra-Corp Technol 2001;33:100–105. 8. Zingg W. Accidental hypothermia. Med Serv J Can 1966;22:399–410. 9. Walsh WB, Betz NE. Tests and Assessment. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall;1990. 10. Roid GH. Stanford-Binet Intelligence Scales. 5th ed. Boston: Riverside Publication Co. (Houghton-Mifflin), 2003. Bender L. Bender Visual Motor Gestalt. New York: Psychological Corporation, 1938. 11. Brunet O, Lezine I. L’development psychologique de la premiere enfant. Paris: Press Universitaires de France;1951. 12. Dragovich D, Tamburlini G, Alisjahbana A, Kambarami R, Karagulova J, Licetto O, Malla DS, Mello MJ, Vani NS. Thermal control of
538
13. 14. 15.
16. 17.
18.
19.
20.
21. 22.
Culic / Archives of Medical Research 36 (2005) 532–538 the newborn. Knowledge and practice of health professional in seven countries. Acta Pediatr 1997;86:645–650. Nielsen JD. Thrombogenesis. Ann Chir Gynecol 1995;84:327–334. Levi M. Pathogenesis and treatment of disseminated intravascular coagulation in the septic patient. J Crit Care 2001;16:167–177. Russwurm S, Stonans I, Schwerter K, Stonane E, Meissner W, Reinhart K. Direct influence of mild hypothermia on cytokine expression and release in cultures of human peripheral blood mononuclear cells. J Interferon Cytokine Res 2002;22:215–221. Ao H, Moon JK, Tashiro M, Terasaki H. Delayed platelet dysfunction in prolonged induced canie hypothermia. Resuscitation 2001;51:83–90. Culic S, Jakl R, Metlicic V, Paukovic-Sekulic B, Resic B, Culic V, Kuljis D, Primorac D. Platelet function analysis in children with Schonlein-Henoch syndrome. Arch Med Res 2001;32:268–272. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M, Tamaki S, Deguchi A, Mori Y, Deguchi K, Shirakawa S, Suzuki K. Hemostatic study before onset of disseminated intravascular coagulation. Am J Hematol 1993;43:190–194. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M, Tamaki S, Deguchi K, Shirakawa S, Hayashi T, Suzuki K. Increased vascular endothelial cell markers in patients with disseminated intravascular coagulation. Am J Hematol 1993;44:85–88. Wada H, Tanigawa M, Wakita Y, Nakase T, Minamikawa K, Kaneko T, Ohiwa M, Kageyama S, Kobayashi T, Noguchi T, Deguchi K, Shirakawa S. Increased plasma level of interleukin-6 in disseminated intravascular coagulation. Blood Coagul Fibrinol 1993;4:583–590. Vinazzer HA. Antithrombin III in shock and disseminated intravascular coagulation. Clin Appl Thromb Hemost 1995;1:62–65. Bellinger DC, Jonas RA, Rappaport LA, Wypij D, Wernovsky G, Kuban KCK, Barnes PD, Holmes GL, Hickey PR, Strand RD, Walsh AZ, Helmers SL, Constantinou JE, Carrazana EJ, Mayer JE, Hanley FL, Castaneda AR, Ware JH, Newburger JW. Developmental and neurologic status of children after heart surgery with hypothermic circulatory
23.
24.
25.
26.
27.
28.
29. 30.
arrest or low-flow cardiopulmonary bypass. N Engl J Med 1995; 332:549–555. Miller G, Tesman JR, Ramer JC, Baylen BG, Myers JL. Outcome after open-heart surgery in infants and children. J Child Neurol 1996;11: 49–53. Hall MW, Goodman PD, Alston SM, Solen KA. Hypothermia-induced platelet aggregation in heparinized flowing human blood: identification of a high responder subpopulation. Am J Hematol 2002;69:45–55. Yoshikawa T, Murakami M, Furukawa Y, Takemura S, Kondo M. Prevention by methylprednisolone of disseminated intravascular coagulation induced by sustained infusion of endotoxin in rats. Haemostasis 1983;13:268–273. Yamazaki M, Aoshima K, Mizutani T, Ontachi Y, Saito M, Morishita E, Asakura H, Matsuda T, Triplett DA. Prednisolone inhibits endotoxininduced disseminated intravascular coagulation and improves mortality in rats: importance of inflammatory cytokine suppression. Blood Coagul Fibrinol 1999;10:321–330. Moscardo F, Perez F, de la Rubia J, Balerdi B, Lorenzo JI, Senent ML, Aznar I, Carceller S, Sanz MA. Successful treatment of severe intraabdominal bleeding associated with disseminated intravascular coagulation using recombinant activated factor VII. Br J Haematol 2001; 114:174–176. Chuansumrit A, Chantarojanasiri T, Isarangkura P, Teeraratkul S, Hongeng S, Hathirat P. Recombinant activated factor VII in children with acute bleeding resulting from liver failure and disseminated intravascular coagulation. Blood Coagul Fibrinol 2000;11(Suppl 1): S101–S105. Morley D, Blumenthal I. A neonatal hypothermia indicator. Lancet 2000;355:659–660. Goldsmith JR, Arbeli Y, Stone D. Preventability of neonatal cold injury and its contribution to neonatal mortality. Environ Health Perspect 1991;94:55–59.
ORIGINAL ARTICLE
Cold Injury Syndrome and Neurodevelopmental Changes in Survivors Srdjana Culic Department of Pediatrics, Clinical Hospital Split, Split, Spinciceva, Croatia Received for publication July 23, 2004; accepted March 18, 2005 (ARCMED-D-04-00041).
Background. The aim of this study was to extend and develop knowledge of the clinical parameters of cold injury syndrome (CIS) in hypothermic infants and newborns, so that morbidity, mortality, and neurodevelopmental disturbances associated with the condition can be reduced. Methods. This retrospective, 10-year cohort study investigated 103 hypothermic infant and newborn patients (aged 0–60 days) admitted to the Pediatric Department between January 1, 1976 and December 31, 1985. Medical records of 67 survivors and 36 children who died were analyzed. Of the survivors, 24 consented to undergo assessment of neuropsychological and neurodevelopmental outcomes. Results. At presentation, all patients manifested cold skin and limbs, while 96 (93.2%) were sleepy, hypotonic and inactive. Cyanosis with bradycardia, bradypnea, and apneic crisis (AC) was observed in 80 (77.7%) patients, and 65 (63.1%) demonstrated diffuse intravascular coagulation (DIC). Pulmonary hemorrhage, AC, respiratory distress syndrome, and DIC represented the primary causes of death. Higher body temperatures on admission reduced the risk of death. Of the 24 patients assessed for neuropsychological and neurodevelopmental outcomes, all but one had normal intellectual capabilities, whereas 16 (66.7%) showed signs of neurodevelopmental disturbances. Conclusions. The severity of hypothermia, leukopenia, and thrombocytopenia with DIC correlates with prognosis and death rate. Hypothermia in infants and newborns can cause psychological and neurodevelopmental disturbances in survivors. 쑖 2005 IMSS. Published by Elsevier Inc. Key Words: Hypothermia, Infants, Diffuse intravascular coagulation, Neurodevelopment.
Introduction Hypothermia is defined as a core body temperature (BT) of 35⬚C or lower. It can be classified according to one of the following categories: mild (BT: 32.2–35⬚C); moderate (BT: 28–32.2⬚C); and severe (BT: ⬍28⬚C). Due to the inadequacy of standardized thermometers, hypothermia frequently remains unrecognized as its accurate diagnosis requires a low-reading thermometer. The hypothermia syndrome is frequently life-threatening and is characterized by multiple Address reprint requests to: Srdjana Culic, PhD, Clinical Hospital Split, Pediatrics Clinic, 21000 Split, Spinciceva 1, Croatia; E-mail: srdjana. [email protected]
0188-4409/05 $–see front matter. Copyright d o i : 10 .1 0 1 6/ j.ar c med .2 0 05 .0 3 .0 4 3
causative factors and a diverse spectrum of pathophysiological consequences. Cold injury syndrome (CIS) may be described as a type of accidental hypothermia recognized in our region (Croatia) and in Israel (1,2). The term cold injury syndrome was first used by Mann at the Royal Society of Medicine meeting in 1953 to describe the condition observed in pre-term, newborn infants exposed to cold (2–4). Between 1955 and 1957, Mann and Elliot (4) and Smallpiece and Jones (5) published a great deal of new information on CIS. From reviews of this and other reports in the literature, it is evident that CIS is a unique clinical entity occurring in infants and neonates with hypothermia.
쑖 2005 IMSS. Published by Elsevier Inc.
Cold Injury Syndrome
Interview protocols for the assessment of psychological and neurodevelopmental outcomes in the 24 survivors who consented to further assessment included evaluation of both intellectual capabilities and signs of neurological dysfunction. For children ⬎4 years, the Kohs test (9) and the Binet⫺ Simone scale (10) were used and visual–motor perception was assessed using the Bender–Gestalt test (11). In order to evaluate psychological development in children ⬍4 years of age, the Brunet–Lezine scale (11) was applied. In addition, a psychological interview was carried out by the hospital psychologist. Following psychological and neurodevelopmental evaluation, the 24 survivors were categorized into one of three groups: 1) patients with normal psychological functioning; 2) patients with normal intellectual development and capabilities, but showing signs of neurological disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual perception, and other specific cognitive deficits and developmental retardation, e.g., enuresis; 3) patients with lower intellectual capabilities.
Results From January 1, 1976 to December 31, 1985, the overall number of hospitalized children was 17,753, with a total of 266 (1.5%) deaths. CIS was diagnosed in 103 children, and 36 (34.95%) of these patients died. We calculate that the frequency of this disease during the period between January 1976 and December 1985 was 0.58% with a mortality rate of 13.53%, which represents a significant proportion of patients. Children suffering from CIS were most often newborn babies; 80 (77.67%) patients were between 0 and 15 days old. Figure 1 shows the ages at admittance of those patients
50 45
Children suffering from CIS
40 35
Children dying from CIS
30 25 20 15 10 5
-5
5
5 41
-4
5 31
-3
5 -2 21
11
-1
5
0 5
The databases of the Pediatric Department of the Clinical Hospital in Split were searched for patients diagnosed with CIS between January 1, 1976 and December 31, 1985. During this period, 103 children with signs of both hypothermia and subsequent organ injuries presented and were treated at the Pediatric Department. None of the CIS patients had coexisting or predisposing conditions. The clinical and laboratory records of these patients were retrospectively reviewed. Additional data were collected from 24 CIS survivors who responded, or whose parents responded, to an initial telephone poll and consented to undergo psychological and neurodevelopmental evaluation.
0-
Subjects and Methods
51
Severe hypothermia causes multiple systemic disturbances that in turn may lead to decreased tissue oxygenation, circulatory collapse, hepatic and renal failure, dehydration, and inappropriate activation of the coagulation cascade; the latter may precipitate diffuse intravascular coagulation (DIC) (6). Decreased oxygen consumption and metabolic rate may precede hypothermia-induced endothelial damage and subsequent consumption coagulopathy. Investigating the effects of temperature change on the coagulation time of blood, Kmiecik et al. (7) reported that activated clotting time increases as body temperature decreases from 37⬚ to 27⬚C. Severe hypothermia produces cardiac complications, apnea, fixed and dilated pupils, and an isoelectric EEG, and attempts to resuscitate the patient by increasing body temperature often fail (8). Infants with CIS cool slowly and are usually hospitalized before such complications occur. Where decrease in body temperature occurs rapidly to ⬍30⬚C, organ ischemia may be avoided. The current study supports the hypothesis that CIS is a distinct clinical entity, showing patterns of national, and particularly regional, pathology. Analyses performed during this study are based on documentation from infants treated for CIS at the Pediatric Department of the Clinical Hospital in Split between 1976 and 1985. Data were collected from patient records and from telephone polls completed by those CIS survivors who agreed to provide information. The aim of the study was to evaluate 1) all patients with CIS who were treated in the Pediatric Department at the Clinical Hospital in Split during the 10-year period from January 1, 1976 to December 31, 1985; 2) the exogenous factors that trigger the onset of the syndrome; and 3) the neurodevelopmental outcome of CIS survivors. We hypothesized that hypothermia may induce neurodevelopmental disturbances, including one or more of the following symptoms: deficits in attention, concentration, visual–motor perception; hyperactivities; disturbances in motor coordination; altered cognitive functioning; and developmental retardation. In addition, we sought to analyze the pattern of hypothermia, its connection with DIC, and its effects on organ systems in neonates and infants.
533
Figure 1. Age (days) at admittance of patients with CIS.
534
Culic / Archives of Medical Research 36 (2005) 532–538
who suffered and/or died from the condition. There were 61 (59.22%) boys and 42 (40.78%) girls; 47 (45.63%) of these babies were full-term, whereas 45 (43.69%) were pre-term. In addition, 86 (83.49%) patients had a birth weight of ⬎2500 g. A total of 78 (75.73%) patients came from the village regions; of these, 28 (27.18%) and 22 (21.36%) patients were from the surrounding regions of the towns of Sinj and Split, respectively, whereas 12 (11.65%) patients came from the town of Split itself. Children were hospitalized more frequently in the winter than in the summer months, with 32 (31.07%) and 21 (20.39%) patients admitted in December and January, respectively. During the summer, there were no reported cases of CIS. In addition, the number of children suffering and/or dying from CIS decreased during the period between 1976 and 1985 (Figure 2). Frequently, CIS patients came to the department in serious condition, with signs of hypothermia and organ injuries already evident. A severe medical condition on admittance was observed in 84 (81.6%) infants, whereas 19 (18.5%) patients were considered to be in moderate condition. All children manifested cold skin and limbs along with red facial skin, whereas 96 (93.2%) were sleepy, hypotonic and inactive. Furthermore, 80 (77.66%) patients were cyanotic with bradycardia and bradypnea, while 77 (74.76%) manifested scleredema. Other symptoms included jaundice, apneic crisis (AC), vomiting, seizures, and secondary infections (Table 1). The lowest rectal body temperature recorded at admission was 21⬚C, although rectal temperatures ranged from 21⬚ to 36⬚C. It was observed that the higher the rectal temperature at admission, the lower the risk of death (Figure 3). According to the anamnestic data, the duration of the condition was very short, often lasting for only 1 day and 23 (63.8%) of the 36 deaths occurrred within the first 24 h
18
Table 1. Symptoms shown by patients with CIS admitted between January 1, 1976 and December 31, 1985 Symptom
Number of patients (%)
Cold skin and extremities Red facial skin Adynamia Lethargy Hypotonia Anorexia Cyanosis Scleredema Bradycardia Bradypnea Jaundice Apnea Vomiting Infection Seizures
103 103 96 96 96 95 80 77 70 69 61 38 30 29 21
(100) (100) (93.2) (93.2) (93.2) (92.2) (77.7) (74.7) (67.9) (66.9) (59.2) (36.8) (29.1) (28.1) (20.3)
(Table 2). Additionally, 20/47 (42.55%) full-term babies died compared with 14/45 (31.12%) pre-term babies (Table 3). Of the 17 patients with a birth weight of ⬍2500 g, 10 (58.82%) died, whereas of 86 patients with a birthweight of ⬎2500 g, 26 (30.23%) died (Table 4). Glycemia was detected in 66 (64.07%) patients, whereas 42 (40.77%) children were found to be hypoglycemic. In 65 (63.10%) CIS patients, DIC was observed, 21 (26.58%) patients had thrombocytopenia with DIC, and 36 (45.57%) demonstrated both thrombocytopenia and leukopenia with DIC (Table 5). In addition, 87 (84.47%) children were polyglobulic. Due to technical reasons, antithrombin III (AT III) levels were analyzed in only four patients at admission; these four patients showed AT III levels of 60%, 34%, 35%, and 24%, respectively. Most (28/36; 72.22%) patients who died as a result of CIS were ⬍15 days old. All children who died had metabolic acidosis and AC due to pulmonary hemorrhage. There were
12
16 10
14 12 10 8
Children suffering from CIS
8
Children dying from CIS
6
Children suffered from CIS Children died from CIS
4
6 4
2
2 0
0 1976
1978
1980
1982
1984
Figure 2. The number of children suffering/dying from CIS between January 1, 1976 and December 31, 1985.
2122
2324
2526
2728
2930
3132
3334
3536
Figure 3. Body rectal temperature (⬚C) at admittance of children suffering/ dying from CIS.
Cold Injury Syndrome
535
Table 2. Length of time to death in CIS patients following admittance
Table 4. Incidence of CIS and death: birth weight
Length of time to death (h)
Birth weight (g)
Number of patients (%)
24 48 72 ⬎72 Total
23 7 1 5 36
(63.8) (19.4) (2.7) (13.8) (100)
no differences in death rates between boys and girls. Seven (87.5%) of eight patients with a temperature ⬍25⬚C died, whereas 50% of children with a temperature between 25.1⬚ and 26⬚C survived (Figure 3). Further analysis of deceased patients revealed that 29 (80.55%) children had a low white blood cell (WBC) count, with 19 (52.78%) children demonstrating a WBC level of ⬍2.0 × 109/L (normal levels are 4–10 × 109/L). Neurodevelopmental assessment in the 24 CIS survivors revealed that 7 (29.17%) patients had had normal neuromotor development, whereas 16 (66.67%) showed normal intellectual capabilities with signs of neurodevelopmental disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual perception, as well as other specific cognitive deficits and developmental retardation (Figure 4). Only one (4.16%) patient, born as a pre-term baby, demonstrated diminished intellectual capacity. Further analyses found no correlation between gestational age and occurrence of neurodevelopmental disturbances. Only one (4.17%) post-term child with normal intellectual capabilities and neurodevelopmental disturbances was observed (Figure 4).
Discussion Dragovich (12) has described how hypothermia is a common problem in neonates and an important contributory factor in neonatal morbidity and mortality, particularly in developing countries where knowledge and practice of the thermal control of newborns is insufficient. Increased awareness and improved knowledge of thermal regulation and thermal protection in such countries may help to prevent CIS in both newborn babies and young infants, whereas improved understanding of its clinical manifestations may hasten diagnosis and aid effective treatment. We report here on the demographic parameters and clinical and laboratory presentations of CIS in neonates and Table 3. Incidence of CIS and death: gestational age
Gestational age
Pre-term (%)
Full-term (%)
Post-term (%)
Total (%)
Incidence of CIS Incidence of deaths
45 (43.6) 14 (38.8)
47 (45.6) 20 (55.5)
11 (10.6) 2 (5.5)
103 (100) 36 (34.9)
1000–1500 1501–2000 2001–2500 2501–3000 3001–3500 3501–4000 4001–4500 Total
Number of patients (%)
Number of deaths (%)
1 (0.9) 7 (6.8) 9 (8.7) 45 (43.7) 32 (31.1) 8 (7.8) 1 (0.9) 103
1 (2.8) 5 (13.9) 4 (11.1) 13 (36.1) 9 (25) 4 (11.1) 0 36
infants, as well as potential neurodevelopmental consequences in survivors of this condition. Our study revealed that the death rate of CIS patients was highest in December and November, which are cold months during which patients tended to be particularly severely affected. At the time of admittance the lowest rectal temperature recorded was 21⬚C, and rectal temperatures ranged from 21⬚ to 36⬚C. Higher body temperatures at admittance were associated with a reduced mortality rate; however, one infant with a rectal temperature of 22⬚C did survive. Rate of death was independent of birth weight or gestational age (Table 4). The cause of death in patients with CIS is almost certainly thrombocytopenia with DIC with or without leukopenia, which results from hypothermic stress and shock. Normal hemostasis is a dynamic balance between bleeding and thrombosis (13), and DIC is considered to be a systemic failure of coagulation, leading to consumption coagulopathy and a tendency to bleeding and multiple thrombi causing dysfunction of one or multiple organs. In addition, platelet contact induces a variety of leukocyte reactions and results in the production and secretion of several cytokines and adhesive proteins associated with the inflammatory response. Activated coagulation factors and cytokines such as interleukin 1 (IL-1) or tumor necrosis factor (TNF) also mediate
Table 5. Incidence of death: WBC, PLT, and DIC Number of patients (%) Thrombocytopenia and DIC Thrombocytopenia without DIC Leukopenia and DIC Leukopenia without DIC Thrombocytopenia, leukopenia without DIC Thrombocytopenia, leukopenia, DIC Decrease in thrombocyte numbers without thrombocytopenia, DIC Total
21 3 1 4 7
Number of deaths (%)
(36.70) (3.79) (1.26) (5.26) (9.21)
7 (33.3) 0 0 1 (25) 3 (42.8)
36 (45.56)
22 (61.1)
7 (8.86) 79
0 33 (41.7)
WBC, white blood cells; PLT, platelet numbers; DIC, diffuse intravascular coagulation.
536 16 14 12 10 8 6 4 2
Culic / Archives of Medical Research 36 (2005) 532–538 Normal psychological functioning Normal intellectual capabilities with signs of neurodevelopmental dysfunctions Low intellectual capabilities
Table 6. Incidence of death: platelet levels PLT 109/L
Number of patients (%)
⬎110 110–70 69–30 ⬍30 Total
27 30 29 17 103
Number of deaths (%)
(26.21) (29.12) (28.15) (16.50)
3 12 13 8 36
(11.1) (40.0) (44.8) (47.7) (34.9)
PLT, platelets.
0
Figure 4. Neurodevelopmental outcomes of CIS survivors.
activation of endothelial cells and lead to intravascular fibrin deposition (14). Russwurm et al. (15) have investigated the influence of hypothermia on cytokine expression and concluded that secretion of IL-2, IL-6, IL-10, and TNF-α was reduced in hypothermic cultures of peripheral blood mononuclear cells. The authors suggested that reduction of IL-2 production might impair the immune response leading to an increase in infectious complications. The pathophysiological manifestation of DIC is a dynamic process that occurs in three phases. While in the first phase hemostasis is compensated, in the second phase uncompensated activation of the hemostatic system occurs and is associated with decreased organ function and bleeding from injuries and venous puncture sites. DIC becomes fully expressed during the third phase, when it is characterized by skin bleeds, multi-organ failure, and an extremely prolonged prothrombin time (PT) and activated partial thromboplastin time (aPTT). In addition, secondary fibrinolysis, which inhibits vascular occlusion, induces consumption of platelets, leukocytes, coagulation factors, and inhibitors. During this phase, platelet and leukocyte counts are very low, and coagulation factor activities are ⬍50% lower than normal activity levels. These observations are supported by data from animal studies. Ao and co-workers (16) concluded that, in dogs, long-term hypothermia induces platelet dysfunction leading to decreased platelet aggregation and prolonged coagulation time. We revealed platelet dysfunction by a platelet function analysis with standard platelet aggregation tests that we performed in just one patient. Based on the aggregograms examined by standard Born method (17), in this patient platelet function analysis by ADP as induction substance revealed the block of endogenous ADP release. When we induced platelet aggregation via another stimulus such as adrenalin, we found that aggregation was absent. Technical limitations prohibited further analysis of platelet function in our patients. In the present study, the majority of children (81.5%) presented in serious condition, with signs of hypothermia already expressed and organ injuries probably in the second or third phase of DIC. Analysis of laboratory tests revealed a high frequency of leukopenia and thrombocytopenia, which were generally associated with DIC. Tables 6 and 7 show
the correlation between platelet and white blood cell counts and the incidence of death; these findings suggest that the leukocytes in the microvascular thrombi were consumed. In addition, we noted that in a small percentage of cases selective bone marrow injury caused by hypothermia and producing thrombocytopenia (3 [3.79%] patients) or leukopenia (4 [5.26%] patients) without DIC. Criteria for diagnosis of DIC have yet to be established. Wada et al. (18) observed hemostatic abnormalities prior to the onset of DIC and suggested that the monitoring of the plasma FDP-D-dimer in pre-DIC conditions may be beneficial. They further suggest that the prognosis of patients with DIC may be related to organ failure or endothelial cell damage (19). Imbalances in the cytokine network and its influence on coagulation have also been observed (20). Future investigations of cytokines such as IL-6, TNF-α, and IL-1β in children with CIS may help to explain the underlying pathophysiology of the syndrome. In any case, the degree of leukopenia and thrombocytopenia with the presence of DIC forms a laboratory parameter important for the estimation of prognosis and death rate. AT III is a physiological inhibitor of activated clotting factors. If the available biological activity of AT III diminishes below approximately 50% of normal activity, the inactivation of clotting enzymes is greatly inhibited. In three of four patients with DIC in the current study, AT III levels were significantly diminished. In such cases, it is important to substitute the depleted AT III (21). In the majority of survivors, neurodevelopmental outcome was assessed at 3–10 years of age. We observed that 16 of 24 (66.67 %) patients showed signs of neurodevelopmental disturbances such as hyperactivity, attention deficit disorder, dysfunction of motor coordination and/or visual
Table 7. Incidence of death: white blood cell levels WBC 109/L
Number of patients (%)
⬎5.0 5.0–3.5 3.4–2.0 ⬍2.0 Total WBC, white blood cells.
38 27 17 21 103
(36.89) (26.21) (16.50) (20.38)
Number of deaths (%) 7 3 7 19 36
(18.4) (11.1) (41.1) (90.4) (34.95)
Cold Injury Syndrome
perception, and other specific cognitive deficits and developmental retardation, although intellectual capabilities in these patients were not impaired. Recent studies (22,23) investigated the influence of hypothermic circulatory arrest as a means of supporting vital organs during open-heart surgery in infants. These reports highlight the presence of a strong association between neurodevelopmental abnormalities and the duration of deep hypothermia, although this association was not evident in all cases. Hall et al. (24) investigated cold-induced platelet aggregation in platelet-rich plasma and concluded that platelet aggregates in the blood of donors with a high number of long-lasting aggregates were rigid and occlusive. The authors suggested that such aggregates may contribute to the cognitive dysfunction noted in patients undergoing hypothermic open-heart surgery. These studies indicate, as does our own experience, that the brain is extremely vulnerable to deep hypothermia in early infancy. Optimal treatment of CIS patients was not assessed in this study. However, we suggest that the optimal treatment for CIS is comprised of the re-warming of the patient to combat the underlying condition and the administration of high doses of corticosteroids. Such a regimen targets both RDS and DIC which, together, represent the primary causes of death in CIS patients. As soon as possible after presentation, a deep body (rectal) temperature reading should be obtained and the patient should be placed under a preheated radiant heater or inside a closed incubator. Active external rewarming is usually sufficient if the core temperature is ⬎32⬚C. However, active core rewarming should also be performed in patients with a core temperature of ⱕ32⬚C. The current trend in active external re-warming is to rewarm slowly. Our recommendation is to re-warm at a rate of 1⬚C/h, starting at 29⬚C, using a radiant heater or a closed incubator. Core re-warming involves the use of warmed (37⬚C) intravenous infusions and extracorporeal blood warming. The patient’s vital signs should be frequently monitored by personnel trained in pediatric intensive care. Whether the rate of re-warming is related to survival rate in hypothermic infants remains controversial, and there are no data in the literature pertaining to the effects of re-warming rate on CIS. An experimental model of DIC used to assess the preventive effects of methylprednisolone suggests that this drug may be beneficial in the treatment of conditions that induce DIC, such as CIS (25). Yamazaki et al. (26) have shown that prednisolone is more effective than heparin in reducing mortality in DIC patients. Our recommendation for the treatment of hypothermic infants is comprised of the use of high doses of methylprednisolone, low-molecular-weight heparin and antithrombin III, as well as re-warming. One of the more recently available treatment options for DIC is recombinant acivated factor VII (rFVIIa, NovoSeven), which has shown efficacy in controlling the acute bleeding in adults and children that results from the consumption of clotting factors in the later phases of DIC (27,28).
537
The frequency of reports of CIS in the hospital database reflects behavioral customs and insufficient home heating to which many infants at this time were subject. Even now, despite warmer homes and changes in midwifery practice, infants are still being admitted for hypothermia in our region. For example, in the period between January 1986 and December 1995, 29 infants with CIS were treated in our clinic. Of these, nine (31%) children died. In contrast, between 1991 and 1995, only five infants were hospitalized with CIS and all survived. This suggests that, although cases of CIS still occur, the incidence of the syndrome is decreasing. Some authors (29) recommend that the Neonatal Hypothermia Indicator should be used in order to prevent hypothermia and CIS in infants. Additional strategies for preventing CIS include parent education and sufficient home heating. Early diagnosis, appropriate re-warming and urgent administration of therapy may prevent psychological and neurodevelopmental disturbances associated with CIS. As Goldsmith and colleagues (30) suggest, excess neonatal mortality during winter months, particularly those deaths attributed to pneumonia or sudden infant death syndrome (SIDS), may be an indicator of missed CIS. In conclusion, the degree of leukopenia and thrombocytopenia in CIS, together with the presence of DIC, forms an important laboratory parameter for the estimation of prognosis and death rate. Early diagnosis is essential for the treatment of DIC. We conclude that intellectual development in children with CIS is generally within the normal range, although neurodevelopmental abnormalities may occur as a result of cerebral hypothermia.
References 1. Stambuk R. Clinical, pathoanatomical features and therapy for cold injury syndrome in children. Ph.D. thesis, Zagreb, Croatia, 1974. 2. Cohen IJ, Amir J, Gedaliah A, Rachmal A, Gorodischer R, Zaizov R. Thrombocytopenia of neonatal cold injury. J Pediatr 1984;104: 620–622. 3. Mann T. Hypothermia in the newborn: a new syndrome. Lancet 1955;2:613–614. 4. Mann T, Elliot RIK. Neonatal cold injury due to accidental exposure to cold. Lancet 1957;1:229–234. 5. Smallpiece W, Jones E. Hypothermia in the newborn. Lancet 1955;1:22. 6. Lazar HL. Treatment of hypothermia. N Engl J Med 1997;337: 1545–1547. 7. Kmiecik SA, Liu JL, Vaadia TS, Nichols JD, Kohtz RJ, Mills NJ, Petterson CM, Stammers AH. Quantitative evaluation of hypothermia, hyperthermia, and hemodilution on coagulation. J Extra-Corp Technol 2001;33:100–105. 8. Zingg W. Accidental hypothermia. Med Serv J Can 1966;22:399–410. 9. Walsh WB, Betz NE. Tests and Assessment. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall;1990. 10. Roid GH. Stanford-Binet Intelligence Scales. 5th ed. Boston: Riverside Publication Co. (Houghton-Mifflin), 2003. Bender L. Bender Visual Motor Gestalt. New York: Psychological Corporation, 1938. 11. Brunet O, Lezine I. L’development psychologique de la premiere enfant. Paris: Press Universitaires de France;1951. 12. Dragovich D, Tamburlini G, Alisjahbana A, Kambarami R, Karagulova J, Licetto O, Malla DS, Mello MJ, Vani NS. Thermal control of
538
13. 14. 15.
16. 17.
18.
19.
20.
21. 22.
Culic / Archives of Medical Research 36 (2005) 532–538 the newborn. Knowledge and practice of health professional in seven countries. Acta Pediatr 1997;86:645–650. Nielsen JD. Thrombogenesis. Ann Chir Gynecol 1995;84:327–334. Levi M. Pathogenesis and treatment of disseminated intravascular coagulation in the septic patient. J Crit Care 2001;16:167–177. Russwurm S, Stonans I, Schwerter K, Stonane E, Meissner W, Reinhart K. Direct influence of mild hypothermia on cytokine expression and release in cultures of human peripheral blood mononuclear cells. J Interferon Cytokine Res 2002;22:215–221. Ao H, Moon JK, Tashiro M, Terasaki H. Delayed platelet dysfunction in prolonged induced canie hypothermia. Resuscitation 2001;51:83–90. Culic S, Jakl R, Metlicic V, Paukovic-Sekulic B, Resic B, Culic V, Kuljis D, Primorac D. Platelet function analysis in children with Schonlein-Henoch syndrome. Arch Med Res 2001;32:268–272. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M, Tamaki S, Deguchi A, Mori Y, Deguchi K, Shirakawa S, Suzuki K. Hemostatic study before onset of disseminated intravascular coagulation. Am J Hematol 1993;43:190–194. Wada H, Minamikawa K, Wakita Y, Nakase T, Kaneko T, Ohiwa M, Tamaki S, Deguchi K, Shirakawa S, Hayashi T, Suzuki K. Increased vascular endothelial cell markers in patients with disseminated intravascular coagulation. Am J Hematol 1993;44:85–88. Wada H, Tanigawa M, Wakita Y, Nakase T, Minamikawa K, Kaneko T, Ohiwa M, Kageyama S, Kobayashi T, Noguchi T, Deguchi K, Shirakawa S. Increased plasma level of interleukin-6 in disseminated intravascular coagulation. Blood Coagul Fibrinol 1993;4:583–590. Vinazzer HA. Antithrombin III in shock and disseminated intravascular coagulation. Clin Appl Thromb Hemost 1995;1:62–65. Bellinger DC, Jonas RA, Rappaport LA, Wypij D, Wernovsky G, Kuban KCK, Barnes PD, Holmes GL, Hickey PR, Strand RD, Walsh AZ, Helmers SL, Constantinou JE, Carrazana EJ, Mayer JE, Hanley FL, Castaneda AR, Ware JH, Newburger JW. Developmental and neurologic status of children after heart surgery with hypothermic circulatory
23.
24.
25.
26.
27.
28.
29. 30.
arrest or low-flow cardiopulmonary bypass. N Engl J Med 1995; 332:549–555. Miller G, Tesman JR, Ramer JC, Baylen BG, Myers JL. Outcome after open-heart surgery in infants and children. J Child Neurol 1996;11: 49–53. Hall MW, Goodman PD, Alston SM, Solen KA. Hypothermia-induced platelet aggregation in heparinized flowing human blood: identification of a high responder subpopulation. Am J Hematol 2002;69:45–55. Yoshikawa T, Murakami M, Furukawa Y, Takemura S, Kondo M. Prevention by methylprednisolone of disseminated intravascular coagulation induced by sustained infusion of endotoxin in rats. Haemostasis 1983;13:268–273. Yamazaki M, Aoshima K, Mizutani T, Ontachi Y, Saito M, Morishita E, Asakura H, Matsuda T, Triplett DA. Prednisolone inhibits endotoxininduced disseminated intravascular coagulation and improves mortality in rats: importance of inflammatory cytokine suppression. Blood Coagul Fibrinol 1999;10:321–330. Moscardo F, Perez F, de la Rubia J, Balerdi B, Lorenzo JI, Senent ML, Aznar I, Carceller S, Sanz MA. Successful treatment of severe intraabdominal bleeding associated with disseminated intravascular coagulation using recombinant activated factor VII. Br J Haematol 2001; 114:174–176. Chuansumrit A, Chantarojanasiri T, Isarangkura P, Teeraratkul S, Hongeng S, Hathirat P. Recombinant activated factor VII in children with acute bleeding resulting from liver failure and disseminated intravascular coagulation. Blood Coagul Fibrinol 2000;11(Suppl 1): S101–S105. Morley D, Blumenthal I. A neonatal hypothermia indicator. Lancet 2000;355:659–660. Goldsmith JR, Arbeli Y, Stone D. Preventability of neonatal cold injury and its contribution to neonatal mortality. Environ Health Perspect 1991;94:55–59.