P&S Medical Review: Nov 1993, Vol.1, No.1
A 28-year-old Black Man with Painful Sickle Cell Crisis

(Clinicopathological Conference)

Michael J. Flamm, M.D.
Charles C. Marboe, M.D.
Abbie I. Knowlton, M.D.
John H. M. Austin, M.D.

Case Presentation

This 28-year-old African-American man with a long history of sickle cell disease entered complaining of "typical" crisis pain in his back.

At age four he was diagnosed with sickle cell disease (SS) when he was admitted to the hospital with back pain, and a temperature of 103 degrees Fahrenheit. Hemoglobin at that time was 7.6 gm/dL with 11.8% reticulocytes. Over the ensuing eight years he had 10 episodes of crisis which brought him to the hospital, requiring admission on two occasions. During this period his hemoglobin varied from 7.2 to 8.3; hematocrit remained in the mid-twenties with 11-18% reticulocytes.

At age 11 he had RUQ pain and gallstones were demonstrated by x-ray. After three exchange transfusions, which decreased the percentage of hemoglobin S from over 80% to 17%, he underwent cholecystectomy and appendectomy without difficulty. Many small calcium bilirubinate stones were found.

Following this operation he did well. He had no further crises until age 21, when he was seen in emergency with severe back pain, a temperature of 99, blood pressure of 132/84, sedimentation rate of 2, white blood cell count of 19.1, hemoglobin of 2 gm/dl, and 21.5% reticulocytes. Chest x-ray showed concave deformities of the end plates of several thoracic vertebral bodies and clear lung fields. He took oral fluids and was given meperidine 75 mg with phenergen 25 mg three times over the next four hours. He was then sent home with tylenol and codeine.

He subsequently moved to the Bronx, where he received care at Bronx-Lebanon Hospital. He reported that he had had crises only rarely and that these were controlled in the Emergency Room with codeine. His most recent crisis was three years prior to his present one.

He had been in his usual state of health until a day before admission, when he developed sweating, fever, and a cough with small amounts of yellow sputum. He also had abdominal pain, dysuria, and on the day of entry, generalized body pain especially over the spine (his usual "crisis" pattern). Upon presentation to the emergency room he was found to be writhing in pain.

On physical examination, the temperature was 99.8, pulse was 88, and respirations were 24. The blood pressure was 120/80. Examination of the head, eyes, ears, nose, and throat were negative apart from pallor of mucous membranes. His neck was supple without lymphadenopathy. Lungs were clear and heart normal. There was abdominal guarding secondary to back pain but no distention. Bowel sounds were present and no masses were felt. A scar from the gallbladder operation was well healed. There was no clubbing or edema.

Apart from his sickle cell disease, his general health in the past was good. No allergies were recorded. Previous operations and injuries included a cholecystectomy at age 11 and broken nose in a baseball game at age 17. He had had recurrent middle ear infections as a child and two myringotomies at age 7. There had been three E. Coli urinary tract infections between the ages of 12-13. At age 15, he had torsion of the testicle.

His father was disabled and his mother was alive and well. Other family members included five brothers, one with sickle SS disease and five sisters, none with sickle disease. His paternal grandmother had asthma.

LABORATORY DATA:

     HEMATOLOGIC VALUES        SERUM CHEMISTRY    

     WBC            24.7      Sodium         137
     hemoglobin      5.4      Potassium        2.9
     hematocrit     17.3      Chloride       104
     RBC             2.25     Carbon dioxide  22
     Neutrophils    69        BUN              8
     Bands           3        Creatinine       1.0
     Lymphocytes    15        Glucose        124
     Monocytes       6        
     Metamyelocytes  2        


Chest x-ray:   --Mildly increased heart size
               --No pulmonary infiltrates
               --Changes compatible with SS disease of the spine.

In the early afternoon he was sleepy but arousable. His temperature then was 102 degrees. On examination basilar crackles were noted in the lungs, and a 2/6 systolic ejection murmur was heard at the apex. There was no S3. His medications since entry had been:

Time           21:00     23:10     01:30     04:00     06:00
Meperidine     75        100       125       100       100
Vistaril       25                   25                  25


Time           08:00     10:00     12:00     14:00     16:00
Meperidine     100       100       100       100       100
Vistaril                  25                  25  

Discussion

Dr. Abbie Knowlton: We see a fair number of sickle cell cases each year. Usually these patients receive large amounts of analgesics such as meperidine, vigorous hydration, and, after being made comfortable, are discharged home within 2 or 3 days. Every so often, however, as in this case, one of these patients dies. This young man had had a good deal of trouble when he was young but was well for the past 2 to 3 years. On this admisison, he came in with what appeared to be one of his usual crises, received standard therapy, and died within 24 hours. I would like to ask a student representative from each of the preceptor groups to present his or her understanding of the patient's immediate cause of death.

Seth Berk, CC3, for the group of Dr. Gerald Thompson: We believe this individual suffered from medication overdose which compromised respiratory function and resulted in acidosis. Relative hypoxia precipitated or exacerbated a sickle cell crisis, leading to subarachnoid hemorrhage -- the most common intracerebral hemorrhage in this population -- and seizure and the subsequent death of the patient.

Rhea McDonough, CC3, for the group of Dr. Robert Chin: We believe the cause of death was hypoxia, brought on by several factors. One of these was the patient's low hematocrit. He may have had a coexisting infection causing an aplastic crisis. Also leading to hypoxia was the seizure, possibly due to meperidine overdose. Meperidine is metabolized to normeperidine, which may reach seizurogenic levels in the brain. In addition, meperidine can cause respiratory depression and hypoxia, another cause for the seizure. The hypoxia and dropping serum potassium level together resulted in cardiac ischemia and arrest.

Sharon Griffith, CC3, for the group of Dr. Rita Charon: We agree with the assessment of Dr. Chin's group. The patient died from hypoxia. In the setting of sickle cell disease and a low hematocrit, he was likely to have been hypoxic on presentation. Rehydration contributed to a further drop in hematocrit, worsening his hypoxia. We believe he suffered a fatal arrhythmia secondary to significant hypokalemia. Though not the cause of death, his seizure was likely due to the combined effects of hypoxemia and arrhythmia.

Dr. Michael Flamm [*]: As the students have suggested, the main point is that no sickle cell patient presenting in this way should be presumed simply to have a vaso-occlusive event. Other possible etiologies for the presenting complaints -- including infection, cardiac disease, and drug overdose -- must be entertained and ruled out. As in the case of this unfortunate patient, many presentations may appear to be "typical painful crises." Numerous other patients -- also initially labeled as having painful crises -- do not have outcomes this unfortunate but are found to have other significant medical problems requiring attention.

I would like to start with a brief review of sickle cell disease, including its clinical course and pathophysiology, to help explain some of this patient's problems. The first case of sickle cell disease was described by James Herrick in the Archives of Internal Medicine in 1910. [1] A young college student from the West Indies presented with anemia, unusually shaped red blood cells, and leg ulcers. Before this time, these patients were thought to have either malaria or other syndromes. Herrick was the first to describe sickle cell disease as a unique entitiy.

In 1949 Pauling demonstrated that the so-called "sickle hemoglobin" had abnormal electrophoretic mobility. Ingram described the amino acid substitution of valine for glutamic acid. With the unravelling of the DNA coding sequences, it was learned that this substitution is a result of a point mutation in the sixth codon of the beta-globin gene, replacing GAG with GTG. Sickle hemoglobin consists of two unaffected alpha chains and 2 sickle beta chains. Individuals with sickle cell disease are homozygous for the autosomal recessive allele. Sickle cell trait is the heterozygous genotype. Of US blacks, 8% have only the trait, while 1 to 2% have the disease.

This genetic defect has been well described for a long time, but we still do not have an effective cure. Although the molecular biology may be completely known, this knowledge does not necessarily translate into effective treatment. In this way sickle cell disease is a paradigm for other genetic diseases such as cystic fibrosis.

The pathology seen in this condition relates to the altered properties of the hemoglobin molecule. Oxygen release and decreased oxygen tension produce a conformational change in the patient's red blood cells and decrease hemoglobin solubility. Erythrocytes sickle, leading to aggregation. Besides deoxygenation, conditions conducive to sickling include stasis, fever, acidosis, and dehydration. Sickling can create a vicious cycle in the microvasculature. Local hypoxia and acidosis increase sickling, clogging up the microvasculature. This stasis leads to further hypoxemia, increased occlusion, and tissue infarction. The so-called "crescent cells," which can occlude capillaries, are irreversibly sickled. However, additional reversible changes occur in other cells. These may also contribute to the disease's pathophysiology.

I will now discuss some of the conditions with which sickle cell patients present. For numerous molecular reasons, the spectrum of clinical severity in sickle cell patients ranges from a few admissions per year to multiple organ damage and early death. Causes for admission in patients with SS disease include: painful crises, cholecystitis, acute chest syndrome, surgical admissions, aplastic crises, hypersplenism, meningitis, urinary tract infections, and other infections.

Types of infections were well characterized in the 1970s, when children with sickle cell disease were found to have increased susceptibility to encapsulated organisms, probably due to splenic dysfunction. Powars showed that many infectious deaths in these children were due to Streptococcus pneumoniae, probably secondary to functional asplenia. [2] This problem has been all but eradicated through the use of pneumovax vaccine and prophylactic daily penicillin in children. Parvovirus infection in any person with chronic hemolysis and increased red cell turnover causes an aplasia. This is the aplastic crisis, a rare entity. Parvovirus infection normally causes a mild drop in hematocrit, but someone dependent on a reticulocyte count of 10 to 30% will experience this as a significant decrease in hematocrit.

The more common crisis, seen on a daily basis in the emergency room, is the so-called vaso-occlusive or painful crisis. Previously it was felt that no correlation existed between vaso-occlusive crises and end organ damage, except avascular osteonecrosis; however, a recent New England Journal of Medicine study showed that patients with high rates of vaso-occlusive crisis may have increased mortality. [3] The age at presentation for these vaso-oclusive crises follows a bimodal distribution, with the first peak occurring in infancy after the sixth month. The teen years may be relatively stable, characterized by fewer admissions for these crises, but the frequency often increases again in adulthood. Of sickle cell patients, 70% have more than one vaso-occlusive crisis per year. Some have more and longer pain-free intervals, while others are frequently admitted to the hospital. Only 23% of sickle cell patients account for 84% of the admissions for vaso-occlusive crisis. [4] Why some patients come in every day and others every three years is unknown. Variations in individual pain thresholds may partially explain this skewed distribution, but other factors must play a role.

One such variable that has been well studied is fetal hemoglobin level. Along with the beta-globin gene the genome contains two gamma-globin genes which are expressed prenatally and shut off at birth. Hemoglobin produced from the gamma-globin genes has a different oxygen-carrying capacity from adult hemoglobin but has the same qualitative function. One group of sickle cell patients found mainly in Saudi Arabia has elevated levels of fetal hemoglobin. These patients are relatively well protected from the complications of their disease. Higher than normal levels of fetal hemoglobin may be a source of protection for those patients with infrequent crises. This is the basis for the use of new medicines, namely hydroxyurea and butyric acid, to increase fetal hemoglobin levels in patients. The rate of vaso-occlusive crisis correlates inversely with fetal hemoglobin level.

Raising patients' hematocrits above their baseline, however, may not be helpful. Transfusion increases blood viscosity and thus may increase the severity of vaso-occlusion.

Dr. Knowlton: Dr. Flamm, you suggest that we might be doing these patients a disservice by giving them blood. At what hematocrit level do you think the severity of anemia outweighs the risks of transfusion?

Dr. Flamm: Patients may present with hematocrits that are decreased from their usual level and symptoms of congestive heart failure: dyspnea on exertion, rales, significant pedal edema. These patients should be transfused to their "baseline" hematocrit. But it is important not to overtransfuse, and to avoid the increased viscosity which might precipitate a vaso-occlusive crisis.

Dr. Knowlton: So, from one patient to another, you would have different hematocrit levels at which to transfuse.

Dr. Flamm: Right. Fetal hemoglobin and other molecular constituents produce variations in steady state hemoglobin levels among these patients. The hematocrit can vary from 15 to 30%, but the decision to transfuse depends on the patient's hematocrit when he or she is not acutely ill.

Let us now review some of the major end organ problems these patients have and discuss them as they pertain to our patient. Sickle cell patients may develop many types of end organ damage. These effects include cerebrovascular accidents, retinopathy, leg ulcers, renal failure, priapism, lung disease, and cor pulmonale due to microthrombi in the pulmonary vessels. Another consequence of sickle cell pain is drug abuse, which may account for some of the disease's morbidity.

Cerebrovascular disease is a major problem, particularly in the pediatric age group. Of all patients, 8 to 10% will have either a cerebrovascular accident or transient ischemic attack. Both large and small cerebral vessels can be affected. A current program using cerebral large-vessel doppler studies is prospectively screening pediatric and adult patients at risk for cerebrovascular accidents. The goal is to determine whether the therapeutic modality of exchange transfusion is useful in such cases. Cerebrovascular disease is the main clinical indication for chronic exchange transfusion. The procedure involves removal of a volume of the patient's packed cells and replacement with AA blood, in an attempt to lower the level of sickle hemoglobin below 30%. Exchange transfusion is the only therapeutic modality clearly shown to decrease the risk of recurrent stroke. Lifetime treatment may be indicated, but most frequently it is done for 5 to 10 years.

These patients develop microinfarcts in the hypertonic medulla of the kidney, isosthenuria (a concentrating defect), and papillary necrosis. Glomerulosclerosis is also fairly common, and a portion of patients develop end-stage renal disease. This may exacerbate their chronic anemia. Chronic lung disease from recurrent pneumonia and from pulmonary vessel thrombosis can cause pulmonary shunting. Together with the chronic anemia, this may result in high-output heart failure.

The incidence of osteonecrosis or avascular necrosis increases with age and can be particularly debilitating, especially when it affects the hips. Avascular necrosis of the femoral head leads to painful crises, and patient's are often unable to walk. Their quality of life after hip replacement may not be much improved, but this is the only modality available after total destruction of the femoral head.

With chronic hemolysis, increased red cell turnover leads to increased levels of bilirubin pigment. Thus, these patients have gallstones and many develop cholecystitis requiring cholecystectomy. Other problems include retinopathy with microinfarction, and retinal detachment. Leg ulcers also occur, probably due to venous stasis and other mechanisms not well understood. From the case history of this patient, it is clear that he previously suffered from gastrointestinal complications at the least. The cardiopulmonary pathophysiology probably also played a significant role in his demise.

Bacterial infection is the most common cause of death in the pediatric population. Cerebrovascular accidents are next, followed by other causes. In the pediatric age group, deaths due to pneumococcal infection have been greatly reduced by the use of prophylactic penicillin and pneumovax. A London Registry study followed 384 sickle cell patients (two-thirds SS, one-third SC) from the mid-1970s to the end of 1989. [5] Out of 18 deaths, 10 were sudden, within 24 hours of admission, like our patient; 2 had strokes; 1 had pulmonary embolism; 1 had pneumococcal sepsis; and half had acute chest syndrome. The latter, which I will now address, is insidious and can be devastating if not recognized immediately. Discussing acute chest syndrome may help us understand some of the factors contributing to this patient's death. The student discussants mentioned several of these causes.

The syndrome of acute chest in adults includes pleuritic chest pain, productive or non-productive cough, shortness of breath, and generally low-grade fever. A new infiltrate may be seen on chest X-ray . If the patient is dehydrated, the infiltrate may not become apparent until after hydration, one or two days after admission. Often no infectious etiology is detectable upon admission.

Acute chest syndrome is a common cause of admission for patients with sickle cell disease. In a 1979 review of lung autopsies of 36 patients who died at Johns Hopkins Hospital, half had thromboemboli, and 8 actually had pulmonary embolisms.[6] This finding suggests that infection may not be the only contributing factor in adult acute chest syndrome. In 1987 Haupt sutdied 74 patients with sickle cell disease and acute chest syndrome. One third had thromboemboli, fewer had evidence of pneumonia, and a significant portion had necrotic bone marrow emboli, an unusual, but potentially catastrophic presentation in the sickle cell patient. [7] Vessel infarction leads to necrosis of the marrow, the same cause of avascular necrosis. Fat particles can be released when the marrow collapses, and these can embolize to the lung, brain, and kidney, contributing to significant mortality.

A 1986 review by Buchanan of 100 patients of various ages admitted with presumed acute chest syndrome, only 2% had blood cultures positive for pneumococcus. [8] Twenty percent had negative initial chest X-rays, but all follow-up X-rays had new infiltrates, and a quarter of these actually developed acute chest syndrome while they were in the hospital. Pulmonary vascular occlusion, hypoxia in the lung, with a PO2 in the 40's, leads to sickling and thrombosis in situ. Bone marrow infarction may release fat particles that embolize to the lungs as well as to the brain and kidney. Infections, such as pneumonia, may present as acute chest syndrome. Most patients presenting with acute chest syndrome should be treated for infection, although frequently no specific organism can be isolated. Patients with streptococcal, Haemophilus influenzae, and Mycoplasma pneumonia, and a large number with Chlamydia or chlamydia-related infections may present with acute chest syndrome.

Now let us review our patient's clinical presentation. We know that he was relatively well until age 21, despite gallstones at age 11, and that he had several admissions after age 21 for painful crises. The day before admission he developed sweating, fever, cough, and chest pain. These symptoms certainly are consistent with the entity of acute chest syndrome. He had abdominal pain, which may have been related to sickle cell disease. No presentations of abdominal pain should be written off as painful crisis. Patients must be evaluated to ensure that they have bowel sounds and a non-tender abdomen. It is essential to know about prior surgery or a history of gallstones. Rare cases of intestinal infarction exist. This patient also had dysuria; we do not know if he had a urinary tract infection. Generalized body pain usually goes along with a crisis.

When seen in the emergency room, the patient was writhing in pain, but he was afebrile and his blood pressure was stable. His respiratory rate was increased, but this could simply have been due to anxiety and pain. His neck was supple so we assume he did not have meningitis. There was no S3 gallop on exam or other evidence of heart failure. The lungs were clear and there was no evidence of fluid overload; but he was dehydrated, so rales may not have been apparent. Abdominal guarding can be due to sickle cell pain referred from the back. There was no distention or masses and bowel sounds were normal, so we do not think he had an acute abdominal process. The cholecystectomy scar was noted. There was no clubbing or edema.

His leukocyte count was 24, 000. We do not know what his baseline count was, nor do we know if his white count had been corrected. The peripheral blood of an individual with a high reticulocyte count may also have many nucleated red cells, squeezed out of the marrow due to increased turnover. Some of these may be counted as white cells. If asked, the lab will correct the white cell count for nucleated reds. The hematocrit on presentation was 17.3. The only other hematocrit we have, from age 21, is 32 with a reticulocyte count of 21% -- if that was his baseline, this presentation represents a significant drop. He may have had white cells in his urine. Gross blood would have been an indication of possible papillary necrosis or another active renal process. His serum electrolytes were remarkable only for a potassium of 2.9 mg/dl, which may represent overhydration and would obviously predispose him to a cardiac arrhythmia.

Radiographic studies of the chest showed mild cardiomegaly, common in sickle cell patients due to their high output state. No pulmonary infiltrate was noted, but as discussed earlier infiltrates in acute chest syndrome can develop a day or two following admission. Also noted were changes consistent with sickle cell disease of the spine due to infarction of the vertebral bodies.

During his hospital course, the patient was given intravenous normal saline and pain-relieving medication. Hypoventilation may have played a role in this patient's demise. Narcotic administration may have decreased respiratory drive. This is not a problem in most patients who have received large amounts of meperidine over their lifetimes because they develop tolerance. If a healthy person received the amount of meperidine that this patient or many other sickle cell patients receive (100 to 125 mg of meperidine every 2 hours), he would probably require naloxone relatively quickly. Because these patients build up tolerance over time, they usually do not develop complications from repeated narcotic administration. Other possible causes include mechanical factors such as rib infarction or, less likely, osteomyelitis of the ribs, or subdiaphragmatic pain due to gallstones or abdominal wall disease, although our patient had his cholecystectomy long before this presentation.

I think we must also consider the possibility of intravascular volume overload contributing to hypoventilation in this patient. We think that a key treatment in sickle cell patients is rehydration. They come in dehydrated, a problem that may be exacerbated by fevers. This intravascular dehydration increases sickling, so rehydration is desirable. However, there is no added benefit to overhydration, and in fact, it may be detrimental One study reported in The New England Journal of Medicine a few years ago, in which patients were made hypotonic (serum sodium between 110 and 120 mg/dl), a large proportion developed seizures. [9] It is desirable just to rehydrate them to normovolemia. This patient may have received too much volume and developed hypokalemia, which may have played a major role in his demise. In addition to possible iatrogenic volume overload, this patient had high-output congestive heart failure, which we know because of his hematocrit of 17 on presentation. We do not know precisely what caused this, but his normal hematocrit was presumably around 30. The much lower than baseline hematocrit suggests that he was in high output congestive failure.

Dr. Knowlton: He was receiving 150 cc/hr. Nothing in the general description states whether he was big or muscular. I do not know his weight and we do not have enough information to be sure of his hydration status.

Dr. Flamm: His grogginess on admission is also significant. He may have been given too much meperidine or he may have had a cerebrovascular accident. A small but demonstrable group of patients use illicit drugs, such as cocaine, and a urine toxicology screen is important. The reason for his grogginess remains unclear.

Student: He had a serum potassium of 2.9 mg/dl. This may have affected his mental status.

Dr. Flamm: In the emergency room, patients often receive hydration for a time, and labs are not drawn until the decision to admit is made. He may not have had a low potassium when he initially presented. His mental status probably changed between coming into the emergency room and arriving at the Allen. On admission at the Allen his temperature was 99 degrees Fahrenheit, pulse was 80 beats per minute. His blood pressure was normal. He continued to receive hydration and pain medication. After admission his grogginess persisted, although he was arousable. He developed crackles in the lung fields, perhaps from volume overload (still no S3 gallop was heard), or from atelectasis secondary to hypoventilation or narcotic overdose. The case summary includes a tabulation of meperidine doses over his course. For a known patient with frequent admissions to CPMC for painful vaso-occlusive crises, this is not an inordinate amount of meperidine; however, since he was never before seen in this hospital, it is unclear how much meperidine this patient normally required.

The cause of seizures in someone taking meperidine is not meperidine but its metabolite, normeperidine, which is renally excreted. Obviously, meperidine is contraindicated in patients with renal insufficiency or focal seizures. The cumulative dose over time contributes to the seizures. We do not have a blood test for normeperidine levels. The metabolite's effect on the CNS may range from twitching or irritability to grand mal seizure. Patients who begin twitching while receiving meperidine should have the medication stopped right away and be closely observed. Even after discontinuing the drug, normeperidine can still accumulate, causing seizures up to a day later. This patient's progression to seizure is a bit rapid for that expected with excessive meperidine, but it can happen; everyone's pharmacokinetics are different.

Student: He received naloxone immediately prior to the seizure. Shouldn't this have helped?

Dr. Flamm: Naloxone only improves mental status and hypoventilation. It does not affect the seizure threshold. I have never seen a normeperidine seizure arrested by naloxone administration. The patient also may have had an arrhythmia, causing brain hypoxia and seizure.

Dr. Knowlton: What was the immediate cause of death?

Dr. Flamm: I think the contributing factors were decreased hematocrit and possibly fluid overload, leading to hypoxia in the probable setting of acute chest syndrome.

Charles Marboe: The pathology showed a number of findings characteristic of sickle cell disease. The most striking finding was autosplenectomy or progressive shrinkage of the spleen over the years until only fibrous tissue remained. Normal spleens are 15 to 18 cm in length and weigh 125 grams. This spleen was about 10 g and measured only 6 or 7 cm in length (figure 1). The microscopic section revealed blue and yellow aggregates or Gandy-Gamna bodies, which contain calcium salts and iron deposits (figure 2). Microthrombosis and hemorrhage of the spleen are the result of repeated sickling.

Ther was no evidence of infection in the patient. Four blood cultures drawn pre and post mortem were negative, and there was no histological focus of inflammation. In the small vessels of the lung there are sickled cells, which are slightly elongated and "banana shaped." Dr. Flamm discussed the predisposing factors to thrombosis. The changes in the lungs are not necessarily thromboemboli, as we did not find a large pulmonary embolus, but the process was diffuse throughout both lungs (figure 3).

Dr. Knowlton: Can you date the process?

Dr. Marboe: It is probably very acute since there is no apparent organization. The alveolar capillaries are clogged with the homogeneous little pink bodies. Due to the extensive and diffuse lung thrombosis, the heart could no longer pump blood effectively through the lungs. The brain shows only red cell sickling. It did not show a subarachnoid hemorrhage, and I suspect his seizure may well have been due to hypoxia.

There are some additional findings expected in sickle cell disease. The liver revealed sickling within the sinusoids. The iron stained specimen revealed a small amount of iron uptake in Kupfer cells. Grossly there was no papillary necrosis in the kidney, but microscopically there was red cell sickling and a small amount of hemosiderosis in the renal tubular epithelium. The bone marrow was not aplastic but was rather reactive. The heart revealed no major infarct or arterial disease, with the exception of a little focus of scar in the papillary muscle of the left ventricle.

In summary, this is an excellent example of the pathological findings expected in acute chest syndrome, in which diffuse thrombosis in the pulmonary microvasculature virtually blocks blood flow through the lungs.

Dr. Flamm: In someone with acute chest syndrome who is rapidly deteriorating (worsening blood gas parameters, adult respiratory distress syndrome requiring intubation, new infiltrate on chest X-ray), the best treatment is exchange transfusion. This patient should have been hypertransfused with additional AA blood at the onset, since his hematocrit was far below baseline. He may not have been able to receive enough to save him. Unfortunately, giving blood to bring his level of AA to 30% might have raised his hematocrit enough to increase blood viscosity. To prevent this, blood should also have been removed using a pheresis machine.

Student: Would obtaining a measurement of arterial blood gases have helped?

Dr. Flamm: Yes. It would have made the situation obvious. Looking at the lung pathology, his hypoxia was the major cause of his demise. A very good review article in the March, 1993 issue of Blood addressed the indications for simple and exchange transfusions. [10]

REFERENCES

1. Herrick, JB. Peculiar elongation and sickle-shaped red blood corpuscles in a case of severe anemia. Arch Intern Med 1910;6:517-21.

2. Powers, D., Overturf, G., Weiss, J., Lee, S., Lhan, L. Pneumococcal septicemia in children with sickle cell anemia. J Am Med Assoc 1981;245:1839-42.

3. Platt, OS, Thorington, BD, Brambilla, DJ, et al. Pain in sickle cell disease. Rates and risk factors. N Engl J Med 1991;325:11-6.

4. Baum, KF, Dunn, DT, Maude, GH, Serjeant, GR. The painful crisis of homozygous sickle cell disease. A study of risk factors. Arch Intern Med 1987;147:1231-4.

5. Gray, A, Anionwu, EN, Davies, SC, Brozovic, M. Patterns of mortality in sickle cell disease in the United Kingdom. J Clin Pathol 1991;44:459-63.

6. Charache, S, Scott, JC, Charache, P. 'Acute chest syndrome' in adults with sickle cell anemia. Arch Intern Med 1979;139:67-9.

7. Haupt, HM, Moore, GW, Bauer, TW, Hutchins, GM. The lung in sickle cell disease. Chest 1987;81:332-7.

8. Buchanan, GR, Smith, SJ. Pneumococcal septicemia dispite pneumococcal vaccine and prescription of penicillin prophylaxis in children with sickle cell anemia. Am J Dis Child 1986;140:428-32.

9. Rosa, RM, Bierer, BE, Thomas, R, et al. A study of induced hyponatremia in the prevention and treatment of sickle-cell crisis. N Engl J Med 1980;303:1138-43.

10. Wayne, AS, Kevy, S, Nathan, DG. Transfusion management of sickle cell disease. Blood 1993;81:1109-1123.