Acute multiorgan failure syndrome: A potentially catastrophic complication of severe sickle cell pain episodes

Acute Multiorgan Failure Syndrome: A Potentially Catastrophic Complication of Severe Sickle Cell Pain Episodes KATHRYN

L.

HASSELL,

M.D.,

JAMES

R.

ECKMAN,

M.D.,

PETER

A.

LANE,

The purpose of this report is to characterize the acute multiorgan failure syndrome that complicates some episodes of sickle pain. A retrospective chart review was used to identify episodes of sickle pain complicated by the acute failure of at least two of three organs: lung, liver, or kidney. The defining criteria of organ failure were established, and the clinical characteristics, laboratory values, treatment methods, and outcomes were noted in episodes that met the criteria. Seventeen episodes of acute multiorgan failure were identified in 14 patients, 10 with sickle cell anemia and 4 with hemoglobin SC disease. Most episodes occurred during a pain event that was unusually severe for the patient. The onset of organ failure was associated with fever, rapid fall in hemoglobin level and platelet count, nonfocal encephalopathy, and rhabdomyolysis. Bacterial cultures were negative in all but four episodes. Aggressive transfusion therapy was associated with survival and with rapid recovery of organ function in all but one episode. The syndrome developed in patients who had previously exhibited relatively mild disease with little evidence of chronic organ damage and relatively high hemoglobin values in steady state.

From the Colorado Sickle Cell Treatment and Research Center (KLH, PAL), and the Departments of Medicine (KLH) and Pediatrics (PAL), University of Colorado School of Medicine, Denver, Colorado, and the Georgia Sickle Cell Center at Grady Memorial Hospital, and the Depart ments of Medicine and Pediatrics, Emory University School of Medicine (JRE), Atlanta, Georgia. This work was supported, in part, by a grant from the National Heart, Lung, and Blood Institute (Grant No. lPGOHL48482). James R. Eckman, MD, was the principal investigator. Requests for reprints should be addressed to Peter A. Lane, MD, Colorado Sickle Cell Treatment and Research Center, Campus Box C-222, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, Colorado 80262. Manuscript submitted March 30, 1993, and accepted in revised form July27, 1993.

February

M.D.,

Denver,

Colorado,

ano~t/ante,

Georgia

Acute multiorgan failure syndrome is a severe, life-threatening complication of pain episodes in patients with otherwise mild sickle cell disease. The syndrome appears to be reversed with prompt, aggressive transfusion therapy. High baseline hemoglobin levels may represent a predisposing factor.

A

cute episodes of vaso-occlusive pain are characteristic of sickle cell disorders and usually are not associated with severe complications. However, some episodes are complicated by acute, life-threatening organ dysfunction such as occurs in acute chest syndrome [1,2], aplastic events [3], splenic sequestration [4,51, and hepatic sequestration with liver dysfunction [61. The acute dysfunction of the lungs or liver has been attributed to sequestration of sickle cells causing microvascular occlusion and tissue ischemia, which may lead to acute organ failure and death [7]. Concurrent failure of multiple organs, during vaso-occlusive pain, has rarely been reported without serious coexisting medical conditions. Of five cases reported with “life-threatening infarctive crises,” one patient had a ruptured spleen, one had salmonella sepsis with abscess, and one .had acute chest syndrome with severe hypoxia [8,9]. The development of severe organ dysfunction and death during painful episodes has been attributed to several causes. Overwhelming infection with sepsis can lead to increased sickling with microvascular occlusion or can directly cause failure of multiple organs from endotoxemia and shock [ 101. The fat embolization syndrome with acute pulmonary and renal failure has been implicated by clinical and autopsy findings in some deaths that occurred during severe pain episodes 111,121. However, in the autopsy series reported by Parfrey et al [13] in 1985, 45% of sickle cell patients who died during a pain episode had no discernible cause of death, suggesting that factors other than infection and fat embolization may cause death in this setting. This report summarizes findings in a series of patients who rapidly developed acute failure of

1994

The American

Journal

of Medicine

Volume

96

155

ACUTE MULTIORGAN FAILURE IN SICKLE CELL DISEASE / HASSELL ET AL

TABLE I Acute Multiorgan Failure Syndrome: Criteria for Organ Insufficiency* Lung Acute pulmonary infiltrate Hypoxia requiring 2 3 liters of oxygen Liver Acute laboratory abnormalitiest ALT level > 5x normal and baseline values Total bilirubin level > 5x normal and baseline values Direct bilirubin level > 2x normal and baseline values Prothrombin time > 3 seconds prolonged Kidney Acute elevation of serum creatinine > 2.0 mg/dL .T = alanineaminotransferase *Syndrome requires documentation of dysfunction of at least two of the three organs. tHepatic dysfunction requires documentation of at least two of the four laboratory criteria.

multiple organs during severe sickle pain episodes. The similarities in clinical presentation at onset, clinical findings as these episodes progressed, and response to therapy define a syndrome that occurs as a complication of sickle cell disease. Because early recognition and aggressive transfusion therapy is associated with survival and a good prognosis for complete recovery, clinicians must be aware of the syndrome when managing patients with sickle pain. PATIENTS

AND METHODS

Definition of Acute Multiorgan

Failure Syndrome

Acute multiorgan failure syndrome is defined as the acute development of severe dysfunction of at least two of three major organs in the setting of a sickle cell pain episode. The criteria used to define dysfunction of three major organs (lung, liver, and kidney) are shown in Table I. Identification

of Patients

All patients with sickle cell disease followed at the Colorado Sickle Cell Treatment and Research Center in Denver and the Georgia Sickle Cell Center at Grady Memorial Hospital in Atlanta were eligible for this retrospective chart review. Of 188 patients currently followed in Denver, 8 episodes of acute multiorgan failure syndrome were identified in 5 patients during a 5-year period. Of 879 patients currently followed in Atlanta, 9 episodes were identified in 9 patients, 8 during 3 years. Other episodes were not included due to insufficient data or incomplete records. Thus, the true incidence and prevalence of this complication could not be determined. Identification Values

of Baseline and Maximum Laboratory

Baseline laboratory values of hemoglobin, platelet count, aspartate aminotransferase CAST), ala156

February 1994

The American Journal of Medicine

nine aminotransferase (ALT), lactate dehydrogenase (LD), total and direct bilirubin, and creatinine were averages of values when patients were seen for routine clinic visits. In cases in which values from clinic visits were not available, laboratory values for direct bilirubin, ALT, and creatinine measured on the day of admission to the hospital, if normal, were used as baseline values. During acute episodes of multiorgan failure, maximum reductions in hemoglobin level and platelet count were defined as the lowest values documented prior to the initiation of transfusion therapy. Absolute reticulocyte counts prior to transfusion were also recorded. Maximum abnormalities of AST, ALT, LD, total and direct bilirubin, and creatinine levels were the highest values recorded during hospitalization. Oxygen saturation was measured by either arterial blood gas or pulse oximetry. Treatment and Outcome Assessment

Intravenous fluid administration and parenteral analgesic administration were recorded. Other supportive measures, including hemodialysis and mechanical ventilation, were tabulated. The characterization of transfusion therapy included defining the type of transfusion (simple versus exchange) and the total number of units used during the episode of acute organ failure. In the case of exchange transfusion, this represented the number of units used during, and within 24 hours after, the exchange. In the case of simple transfusion, this number represented the number of units used during hospitalization. Outcome was assessed by survival, number of days to discharge after the initiation of transfusion therapy, and the length of time to recovery of organ function, defined as return of laboratory values to normal or baseline levels. Statistical Methods

Mean values and standard deviations were calculated for laboratory values at baseline and at the point of maximum change for all patients meeting the definition of acute multiorgan failure syndrome. These means were compared using the paired t-test. RESULTS Characteristics

of the Patients

Seventeen episodes that met the criteria for acute multiorgan failure were identified in 8 females and 6 males. Diagnosis, age, sex, and baseline laboratory values are listed in Table II. The hemoglobin phenotype was documented for each case by cellulose acetate and citrate agar electropho-

Volume 96

ACUTE

MULTIORGAN

FAILURE

IN SICKLE

CELL

DISEASE

/ HASSELL

ET AL

TABLE II Diagnosisand Baseline Laboratory Data in Patients Who DevelopedAcute Multiorgan Failure Syndrome

Patient

Sex

Dx

1

F

SS

2

F

ss

3

F

4

Age

HbF (%I

(Y)

6

Pk

T Bili

D Bili

(x103/1LL)

(mg/dL)

(mg/dL)

ALT N/L)

AST NJ/L)

(mi/rdU

9.8

29.0

487

1.8

0.2

13

NA

0.3

19

8.1

2.2

725

8.8

0.7

NA

58

0.5

ss

21

6.8

NA

475

4.0

NA

NA

99

0.7

M

ss

25

12.0

23.0

445

2.0

0.3

14

33

0.9

5

M

SS

26

9.5

3.1

512

2.6

0.2

27

23

0.8

6

F

ss

30

9.1

2.7

368

2.9

0.5

NA

26

1.1

7

M

ss

33

9.5

4.6

420

3.2

NA

NA

47

0.4

8

M

SS

42

10.6

5.3

350

2.2

NA

13

40

1.1

9

F

ss

49

7.8

7.8

447

2.5

0.1

NA

61

0.9

10

M

ss

59

7.2

9.1

370

3.5

0.4

NA

37

1.2

11

F

SC

24

13.6

3.6

290

1.2

0.5

24

20

0.8

12

M

SC

28

12.6

1.0

176

1.4

0.2

NA

26

1.1

13

F

SC

44

14.5

1.8

450

1.5

30

60

0.8

14

F

SC

61

12.8

1.5

499

0.6

NA

3

0.8

< = diagnosis; SS SC ,.. diseaseitlb = hemo@n; / “-7= sickle cell ,I anemia; ., SC =, hemoglobrn n amlnorransrerase; fib I = asparrare amlnorransrerase; or = creannlne; IUH= nor awaoIe.

NA 0.1

Hb F = fetal hemoglobin; Plt = plateletcount; T Bili = total bilirubin; D BIII = direct bilirubin; ALT = alanine

resis. Ten patients with sickle cell anemia had a median baseline hemoglobin value of 9.3 g/dL (range: 6.8 to 12.0 g/dL), and 4 with hemoglobin SC disease had a median baseline hemoglobin value of 13.2 g/dL (range: 12.6 to 14.5 g/dL). The mean age of patients at presentation of acute multiorgan failure syndrome was 33 years (range: 6 to 61 years). The median age was 28 years. One patient (#4, Table II> had three episodes of acute multiorgan failure, and another patient (#13, Table II) had two episodes. Prior to the onset of acute multiorgan failure, all 14 patients had normal renal and hepatic function. Five patients had a history of previous aeute chest syndrome. Two patients had chronic lung disease (chronic obstruction pulmonary disease, sarcoid), and five had evidence of some chronic tissue damage with avascular necrosis of the femoral head, leg ulceration, proliferative retinopathy, and/or a history of bone infarcts. Eight of the 14 patients had no evidence of chronic organ damage associated with their sickle cell disease. Clinical Presentation All 17 episodes of multiorgan failure occurred in the setting of pain that had a distribution typical of previous pain episodes but that was unusually severe for the patient. In 14 of 17 episodes, initial laboratory values for hepatic and renal function were normal or unFebruary

changed from previous baseline values. These patients were admitted to the hospital for pain management, intravenous hydration, and the administration of analgesics. In nine episodes, supplemental oxygen was administered. Rates of intravenous fluid administration varied from 75 to 200 mL/h. Parenteral pain medications included morphine sulfate via patient-controlled analgesia pump or intramuscular injection, hydromorphone (Dilaudid) intramuscularly, merperidine (Demerol) intramuscularly, buprenorphine (Buprenex) intramuscularly, or nalbuphine (Nubain) intramuscularly; antihistamines (Phenergan, Vistaril) were also given in four cases. In the 14 episodes that developed after admission, onset of organ failure was heralded by sudden clinical deterioration on the third or fourth day of hospitalization, with the development of fever to greater than 38°C and evidence of acute chest syndrome and hepatic and/or renal dysfunction. Confusion and lethargy developed without focal neurologic findings in 12 of these 14 episodes. Bacterial cultures obtained at onset were negative in 11 of 14 episodes. Enterococcal and Escherichia coli urinary tract infections were documented in two patients, and serologic evidence of Erhlichia canis infection of indeterminant onset was detected in a third. Intravenous antibiotics were given during 11 episodes. Evaluation for fat emboli by Sudan red staining of urine and sputum was negative in two episodes. 1994

The American

Journal

of Medicine

Volume

96

157

ACUTE

MULTIORGAN

FAILURE

IN SICKLE

CELL

DISEASE

/ HASSELL

I 700

600

12.0 500 3 5 n’ 0 2. u)

400

300

z a 3 P

200

100

04-

IBaseline Figure counts lowest

0

Pre-Tx

IBaseline

Pre -TX

1. Changes from baseline in hemoglobin levels and platelet during episodes of acute multi-organ failure. Data shown are obtained prior to transfusion therapy. TX = transfusion.

1

ET AL

9.6 + 1.9 g/dL to a mean value of 6.4 +- 1.8 g/dL (p
Evidence of organ failure was present at the time of admission to the hospital in three cases. Onset of pain was 2 to 3 days prior to admission, and fever was present on admission. Nonfocal encephalopathy was noted in one of these three episodes. Bacterial cultures were positive for E. coli in the blood and urine in one case and for Staphylococcus aureus and Pseudomonas sp. in the sputum in another. These patients were treated with intravenous antibiotics.

Acute Pulmonary Failure Acute chest syndrome, defined as the development of an acute infiltrate on chest roentgenograph and hypoxia requiring at least 3 L of supplemental oxygen to maintain an oxygen saturation of at least 90%, occurred in 16 of 17 episodes. Severe chest pain with increasing oxygen requirements progressed to respiratory failure and intubation with mechanical ventilation in four cases. Pulmonary infiltrates as shown by chest roentgenograph were unilateral in 6 cases and bilateral in 10 cases.

Hematologic Changes There was a rapid fall in hemoglobin level in all 17 episodes and a significant fall in platelet count in 16 episodes. Figure 1 shows the change from baseline hemoglobin and platelet values prior to transfusion. In patients with sickle cell anemia, hemoglobin values fell from a mean baseline of

Acute Hepatic Failure Hepatic dysfunction developed in 15 of 17 episodes. Figure 2 shows the ,changes in total bilirubin, direct bilirubin, ALT, ,&ST, and LD levels. The total bilirubin level was measured in all 17 episodes and rose from a mean baseline value of 2.7 + 1.9 mg/dL to a mean maximum value of 13.5 f 14.4

Figure 2. Evidence for hepatic dysfunction during acute multiorgan failure syndrome. Figure shows maximum changes from baseline serum values for total bilirubin, direct bilirubin, alanineaminotransferase, aspartate aminotransferase, and lactate dehydrogenase. Data from all episodes in which these assays were obtained are shown. Lines show changes from baseline for episodes where baseline data were available.

February

1994

The American

Journal

of Medicine

Volume

96

ACUTE

MULTIORGAN

mg/dL (p = 0.005). Direct bilirubin values rose from a mean baseline value of 0.4 ? 0.2 mg/dL to a mean maximum value of 7.2 + 11.7 mg/dL (n = 16, p = 0.05). Prothrombin time was measured in 16 episodes, with a mean maximum value of 18.0 + 5.2 seconds. In 10 of 16, the prothrombin time was prolonged by more than 3 seconds. ALT level rose from normal (mean value: 20 + 6.8 IUIL) to a mean maximum value of 1,720 ? 1,946 IU/L (p = 0.03) in the 11 episodes in which it was measured. AST and LD levels rose from baseline values (20 + 6.9 IU/L and 285 -t 90 IU/L, respectively) to mean maximum values of 1,720 * 1,946 IU/L (n = 16, p = 0.02) and 5,469 ? 5,660 IU/L (n = 13, p = 0.051, respectively. Acute Renal Failure Acute renal insufficiency developed during 13 episodes of acute multiorgan failure syndrome, with a rapid rise in the serum creatinine level during a period of 24 to 36 hours. The serum creatinine level rose from a mean baseline value of 0.8 + 0.2 mg/dL to a mean value of 5.2 + 3.9 mg/dL (p = 0.0003). Oliguria occurred during 11 episodes, and the serum creatinine level was greater than 6.0 mg/dL in 7 episodes. Hyperkalemia, with potassium values greater than 6.0 mEq/L on two consecutive sampies, was present in seven episodes. Three episodes were treated with acute hemodialysis. Rhabdomyolysis The mean maximum creatine phosphokinase (CPK) value measured during 14 episodes was 1,951 IU/L (range: 72 to 7,860 IU/L). Values of greater than 1,000 IUiL occurred during six episodes. Intramuscular injections had been given in four episodes during which the CPK exceeded 1,000 IU/L. Treatment and Outcome All 17 episodes of acute multiorgan failure syndrome were treated with aggressive red cell transfusion as soon as clinical deterioration was recognized. Table III shows the method of transfusion (simple versus exchange) and the mean number of packed red blood cell units given for each episode. Eight episodes were treated with simple transfusion, using a mean of 8 units (range: 4 to 12 units) given over a mean of 7 days (range: 3 to 13 days). Nine episodes were treated with exchange transfusion, eight of these by automated erythrocytopheresis utilizing a mean of 9 packed red blood dell units (range: 8 to 18 units). During this procedure, donor packed red blood cells were reconstituted with the patient’s plasma. Exchange transfusion in February

FAILURE

IN SICKLE

CELL

DISEASE

/ HASSELL

ET AL

TABLE III TransfusionTherapy and Outcomeof Episodesof Acute Multi-Organ Failure Syndrome No.

Units Transfused

Death

Days to Discharge*

8 (4-W+

0

i

9 (8-18)+

1

17 614;;;

Simple transfusion Exchange transfusion *From initiation of tranfuslon therapy. tValues in parentheses are ranges.

the pediatric patient was performed manually with 5 units of packed red blood cells reconstituted with fresh frozen plasma. Post-transfusion hemoglobin electrophoreses documented a mean hemoglobin S or S + C level of 30% (range; 10% to 41%) in five episodes treated with exchange transfusion. Hemoglobin S or S + C levels of 30% and 37% were achieved in two patients treated with simple transfusion (7 and 4 units, respectively). Dramatic clinical improvement was noted within 24 hours of the initiation of transfusion, with clearing of encephalopathy, increases in urine output in the oliguric patients, and decreases in oxygen requirements in all but one patient (#14, Table II). This patient, a 61-year-old woman with hemoglobin SC disease and chronic lung disease, had progressive respiratory failure despite exchange transfusion and died. Autopsy showed pulmonary fat emboli. All other patients showed rapid reversal of organ dysfunction. Table III shows the number of days to discharge after the initiation of transfusion therapy. The eight patients who received simple transfusion for episodes of multiorgan failure syndrome were discharged a mean of 15.5 days after initial transfusion (range: 6 to 27 days), and five patients showed complete recovery of organ function within 3 to 6 months. The only residual laboratory abnormality in the other two patients treated with simple transfusion was an elevated creatinine level in two patients (1.5 and 1.8 mg/dL); no follow-up data were available in the third patient. Of the eight episodes of acute multiorgan failure syndrome successfully treated with exchange transfusion, patients were discharged a mean of 7 days after exchange transfusion (range: 4 to 17 days), and organ function returned to normal within 2 months in all cases.

COMMENTS Acute episodes of pain that occur in patients with sickle cell syndromes are usually self-limited and not associated with severe complications. However, some episodes of pain, particularly those associated with severe infections or other medical conditions, may be complicated by acute, lifethreatening organ dysfunction such as occurs in 1994

The American

Journal

of Medicine

Volume

96

159

ACUTE

MULTIORGAN

FAILURE

i INFECTION

IN SICKLE

.,,’ .. ..._.__.___)

CELL

DISEASE

/ HASSELL

ET AL

DIFFUSE TISSUE ISCHEMIA TRANSFUSION

MULTIORGAN

r

/+

acute chest syndrome [1,2], aplastic events [31, splenic sequestration [4,51, or hepatic sequestration with liver failure [6]. Most of the 17 episodes of acute multiorgan failure reported here occurred in the absence of recognized predisposing conditions. We believe that these cases define a treatable syndrome that complicates sickle cell disease. The failure of multiple organs occurred during a pain episode in which the pain was unusually severe. The organ dysfunction occurred several days into the pain episode, sometimes as the pain was starting to improve. There was a rapid fall in the hemoglobin level and platelet count immediately preceding or coincident with the onset of organ dysfunction. The rapid clinical deterioration was also characterized by fever, usually without a source of infection, and by nonfocal encephalopathy. There was subsequent onset of acute pulmonary, hepatic, and renal failure often with evidence of rhabdomyolysis. All but one patient showed dramatic improvement in organ function after the initiation of aggressive transfusion therapy. Over time, organ function returned essentially to baseline levels. The single fatality occurred in a patient with severe underlying chronic obstructive and restrictive lung disease, who died of respiratory failure despite exchange transfusion and aggressive medical support. The pathophysiology of the acute multiorgan failure syndrome is unclear, but the mechanisms possibly responsible are depicted in Figure 3. Diffuse sickling during severe vaso-occlusive pain episodes with diffuse microvascular occlusion and tissue ischemia is suggested by the simultaneous dysfunction of multiple organs and the develop160

February

1994

The American

Journal

of Medicine

Volume

J

DEATH

Figure 3. Hypothesis of the pathophysiologic mechanisms in acute multiorgan failure occurring during a sickle pain episode. Solid arrows indicate mechanisms common to all cases. Dotted arrows indicate factors that may potentiate organ failure in some cases.

ment of rhabdomyolysis in the majority of cases. The dramatic fall in hemoglobin levels immediately preceding the development of organ failure is consistent with sequestration and destruction of erythrocytes during widespread microvascular occlusion. The exacerbation of anemia would also contribute to tissue ischemia by decreasing the oxygen-carrying capacity. The rapid recovery after large volume simple or exchange transfusion suggests that correction of the acute anemia and/or dilution- of sickling erythrocytes plays an important role in reversing the pathophysiologic mechanisms in this syndrome. Hypoxemia, which could aggravate sickling and vaso-occlusion, may develop from the acute ‘chest syndrome or with hypoventilation from excessive sedation due to narcotic analgesics used to treat the severe pain. In some episodes, increasing hypoxemia and chest pain were documented several hours before severe hepatic and renal dysfunction was recognized. In these cases, hypoxemia may have contributed to diffuse vaso-occlusion and acute organ dysfunction. In other episodes, however, abnormalities in hepatic and renal function occurred concurrently with onset of the acute chest syndrome, suggesting that diffuse vasoocclusion caused simultaneous damage to lungs, liver, and kidneys. Excessive sedation was judged not to be a contributing factor in most cases, in part, because administration of naloxone appeared to have little effect on mental status or oxygen saturation in those with lethargy or coma. Infection could contribute to the development of acute multiorgan failure by worsening microvascular occlusion, by increasing hypoxemia from pneu96

ACUTE

MULTIORGAN

monia, or by causing tissue ischemia from hypotension and endotoxemia associated with sepsis. Fever was present in all episodes of acute multiorgan failure, but bacterial infection was documented in only four episodes. Although most episodes were treated with broad-spectrum antibiotic coverage, antibiotics were not used during three episodes, and these patients showed rapid recovery of organ function after transfusion. No patients were evaluated for viral infections or “atypical” infections, including Chlamydia pneumoniae, Legionella sp. or Mycoplasma sp.; however, these infections do not commonly cause fulminant multiple organ failure. In most of our patients, therefore, the acute multi-organ failure syndrome was not clearly related to overwhelming infection. Fat emboli may occur during sickle pain episodes due to infarction of bone marrow with release of marrow fat and cellular elements into the circulation [12]. The clinical features of fat embolism, especially the fall in hemoglobin level and platelet count, fever, encephalopathy, and acute respiratory and renal failure, are similar to those seen in the episodes of acute multiorgan failure reported here. Fat embolism was documented by autopsy in the only fatal case with persistent respiratory failure despite exchange transfusion. The other 16 episodes of acute multiorgan failure were readily reversed with transfusion therapy. Transfusion with packed red blood cells, without whole blood exchange, would not likely reverse fulminant fat embolism, and, in two of these episodes, fat was not detected in the urine or expectorated sputum. Previously reported cases of fat embolism in sickle cell disease.usually resulted in death 111,141. Thus, although fat embolism may occur during episodes of fatal acute multiorgan failure, the rapid improvement with erythrocytapheresis or with simple transfusion of packed red cells argues against a pathophysiologic role. Two striking clinical characteristics of this group of patients were the relatively high baseline hemoglobin values and relatively benign clinical courses prior to development of acute multiorgan failure. The patients with sickle cell anemia had a median baseline hemoglobin value of 9.3 g/dL, and those with hemoglobin SC disease had a median baseline hemoglobin value of 13.2 g/dL. Higher hemoglobin levels in patients with sickle cell disease have been associated with higher “pain rates” [151. In acute multiorgan failure syndrome, higher baseline hemoglobin levels might be a risk factor because of their contribution to increased blood viscosity. The patients in this series also had relatively few previous episodes of pain and relatively little evidence of chronic organ damage. The occurrence of multiorFebruary

FAILURE

IN SICKLE

CELL

DISEASE

/ HASSELL

ET AL

gan failure in two patients with sickle cell anemia and baseline hemoglobin F levels of 23% and 29%, respectively, is remarkable. Higher levels of hemoglobin F are associated with higher hemoglobin levels but are thought to provide clinical protection in sickle cell disease [16,17]. The higher hemoglobin levels in these patients may have outweighed any benefit provided by higher hemoglobin F levels and predisposed the patients to the development of multiorgan failure during pain episodes. Aggressive transfusion therapy to increase the hemoglobin level and to lower the percentage of cells containing sickle hemoglobin is accepted treatment for some acute complications of sickle cell disease. Aggressive transfusion is widely used for stroke [18] and in some cases of priapism [191, acute chest syndrome 120,211, and hepatic sequestration [22]. Automated or manual exchange transfusion is used to replace patient red blood cells with normal donor red cells. Simple transfusion with red blood cells can accomplish similar results if severe anemia develops from the accelerated destruction or sequestration of sickle cells. The dramatic clinical improvement of severe organ dysfunction after aggressive transfusion in the patients reported herein supports the role of prompt transfusion in preventing death or permanent organ damage in patients with this syndrome. This conclusion is highlighted by two patients who experienced more than one episode of acute multiorgan failure yet had complete reversal of organ damage after each episode. It should also be emphasized that acute multiorgan failure is an unusual complication during an episode of pain and that transfusion is of no value in treating uncomplicated sickle pain episodes 1231. In fact, without a significant fall of the hemoglobin level, simple transfusion may increase blood viscosity and potentiate vasoocclusive complications 1241. We have recognized other episodes of acute multiorgan failure during pain events that were not included in this report because the criteria for organ failure presented here were not documented. Some of these other episodes showed similar clinical characteristics, but transfusion therapy was initiated before the criteria for acute multiorgan failure were met. In addition, a number of other patients at both centers died suddenly during severe vaso-occlusive pain episodes. Their rapid clinical deterioration was characterized by fever, encephalopathy, and respiratory distress, but they died suddenly before abnormalities of liver and renal function were documented and before transfusion therapy was initiated. One such fatal episode occurred in Patient 11 (Table II) 1 year after the episode of acute multiorgan failure reported 1994

The American

Journal

of Medicine

Volume

96

161

ACUTE

MULTIORGAN

FAILURE

IN SICKLE

CELL

DISEASE

/ HASSELL

here. These observations, along with the autopsy data of Parfrey et al [131, suggest that acute multiorgan failure as a consequence of diffuse microvascular occlusion may be an important cause of death during pain episodes. Results reported here also suggest that prompt recognition of this syndrome and the initiation of transfusion therapy can be life-saving.

ACKNOWLEDGMENT We are indebted to Verlyn M. Peterson, MD, for clinical insight and helpful suggestions and to Joyce Howard, PAC, for assistance in locating cases.

REFERENCES

22: 169-74. 7. Ludmerer KM, Kissane JM. Left elbow pain and death in a young woman with sickle-cell anemia. Am J Med 1982; 73: 268-82. 8. Van De Pette JEW, Pearson TC, Slater NGP. Exchange transfusion in life-threateningsicklingcrises. J R Sot Med 1982; 75: 777-80.

February

1994

9. Green M, Hall RJC, Huntsman RG, Lawson A, Pearson T, Wheeler PCG. Sickle cell crisis treated by exchange transfusion treatment of two patients with heterozygous sickle cell syndrome. JAMA 1975; 231: 918-50. 10. Effron MB, Chernow B. Shock. In: Rubinstein E, Faderman DD, editors. Scientific American medicine. New York: Scientific American Inc., 1992; 1: l-12. 11. Chmel H, Bertles JF. Hemoglobin S/C disease in a pregnant woman with crisis and fat embolization syndrome. Am J Med 1975; 58: 563-6. 12. Charache S, Page D. Infarction of bone marrow in the sickle cell disorders. Ann Intern Med 1967; 67: 1195-200. 13. Par-hey NA, Moore GW, Hutchins GM. Is pain crisis a cause of death in sickle cell disease? Am J Clin Pathol 1985; 84: 209-12. 14. Shapiro MP, Hayes JA. Fat embolism in sickle cell disease. Report of a case with brief review of the literature. Arch Intern Med 1984; 144: 181-2. 15. Platt OS, Thorington MS, Brambilla DJ, et al. Pain in sickle cell disease. Rates and risk factors. N Engl J Med 1991; 325: 11-5. 16. Powars DR, Weiss JN, Chan LS, Schroeder WA. Is there a threshold level

1. Davies SC, Lute PJ, Win AA, Riordan JF, Brozovic M. Acutechest syndrome in sickle cell disease. Lancet 1984; 1: 36-8. 2. Sprinkle RH, Cole T, Smith S, Buchanan GR. Acute chest syndrome in children with sickle cell disease: a retrospective analysis of 100 hospitalized cases, Am J Pediatr Hematol Oncol 1986; 8: 105-10. 3. Conrad ME, Studdard H, Anderson LJ. Case report: aplastic crisis in sickle cell disorders: bone marrow necrosis and human parvovirus infection. Am J Med Sci 1986; 295: 212-5. 4. Emond AM, Collis R, Darvill D, Higgs DR, Maude GH, Serjeant GR. Acute splenic sequestration in homozygous sickle cell disease: natural history and management. J Pediatr 1985; 107: 201-6. 5. Bowcock SJ, Nwabucze ED, Cook AE, Aboud HH, Hughes RG. Fatal splenic sequestration in adult sickle cell disease. Clin Lab Haematol 1988; 10: 95-9. 6. Hernandez P, Dorticos E, Espinosa E, Gonzalez X, Svarch E. Clinical features of hepatic sequestration in sickle cell anemia. Haematologia 1989;

162

ET AL

The American

Journal

of Medicine

Volume

of fetal hemoglobin that ameliorates morbidity in sickle cell anemia? Blood 1984; 64: 921-6. 17. Noguchi CT, Rodgers GP, Serjeant G, Schechter AN. Levels of fetal hemoglobin necessary for treatment of sickle cell disease. N Engl J Med 1988; 318:

96-9.

18. Russell MO, Goldberg HI, Reis L, eta/. Transfusion cular abnormalities

therapyforcerebrovasin sickle cell disease. J Pediatr 1976; 88: 382-7.

19. Seeler RA. Intensive transfusion therapy for priapism in boys with sickle cell anemia. J Urol 1973; 110: 360-l. 20. Mallouh AA, Asha M. Beneficial effect of blood transfusion in children with sickle cell chest syndrome. Am J Dis Child 1988; 142: 178-82. 21. Vichinsky E, Lubin BH. Suggested guidelinesfortreatment of children with sickle cell anemia. Hematol Oncol Clin North Am 1987; 1: 483-501. 22. Sheehy TW, Law DE, Wade BH. Exchange transfusion for sickle cell intrahepatic cholestasis. Arch Intern Med 1980; 140: 1364-6. 23. Charache S, Lubin B, Reid CD, editors. Management and therapy of sickle cell disease. (NIH Publication no. 89-2117) 1989: 25-8. 24. Schmalzer EA, Lee JO, Brown AK, Usami S, Chien S. Viscosity of mixtures of sickle and normal red cells at varying hematocrit levels: implications for transfusion. Transfusion 1987; 27: 228-33.

96