Center Logo


Cage S. Johnson
Professor of Medicine
University of Southern California
Director, Comprehensive Sickle Cell Center

The following represents the more or less biased opinions of the author and in no way reflects on the greater wisdom and experience of you, the reader. Your comments are welcome; any ongoing dialogue will be posted on this open forum to be seen by the merely curious as well as those who know you well. Thus it behooves you to observe the Marquis de Queensbury rules in your comments. "As ye sow, so shall ye reap".

The acute chest syndrome (ACS) in sickle cell disease (SCD) can be defined as:

  1. a new infiltrate on chest x-ray
  2. associated with one or more NEW symptoms:
    fever, cough, sputum production, dyspnea, or hypoxia.

The symptom complex may be varied, and not all symptoms are present in every episode; however, some combination of these symptoms is required for this "diagnosis." The ACS is a common complication of the sickling disorders (Hb SS, Hb SC, Hb S ß+-thalassemia, Hb S ß0-thalassemia, etc.) and is responsible for considerable morbidity and mortality in these patients(1,2).

Recent data from the Clinical Course of Sickle Cell Disease Cooperative Study indicate that this complication occurs with an incidence of 10,500/100,000 patients/year(2). ACS occurs most often as a single episode, but certain patients have multiple episodes. A past history of an ACS is associated with early mortality compared to those who have never had an episode. The disorder is most common in the 2 to 4 year age group and gradually declines in incidence with age. It is believed that Hb F exerts a protective effect on those less than 2 years of age. Furthermore, the decline observed in older age groups is partially related to at least two other factors: (a) excess mortality in the group which had an ACS, and (b) fewer viral episodes in adults due to acquired immunity.

The incidence of ACS is related to genotype, where Hb SS = Hb S ß0-thalassemia > Hb SC = Hb S ß+-thalassemia. Additional data showed that a lower hematocrit or a higher Hb F was associated with a reduced incidence, whereas a high WBC was associated with a higher incidence. Other data indicate that recurrent episodes of ACS can lead to chronic lung disease; the resulting hypoxia is associated with the onset of osteonecrosis in multiple sites and with sudden death due to myocardial infarction(3).

The causes of the ACS are multiple; thus, the terminology "ACS" does not indicate a definite diagnosis but reflects the clinical difficulty of defining a distinct etiology in the majority of such episodes. The causes of the ACS include those disorders directly or indirectly related to the sickling process, as well as causes distinct from sickling.

Hemoglobin S related
Unrelated to Hb S
Direct consequences of Hb SIndirect consequences of Hb S
pulmonary infarction - in situ sickling
fat embolism syndrome
hypoventilation secondary to rib/sternal bone infarction
hypoventilation secondary to narcotic administration
pulmonary edema induced by narcotics or fluid overload
fibrin thromboembolism
foreign body/intrinsic bronchial obstruction
opportunistic infection related to HIV-1 infection
acute sarcoidosis
other common pulmonary diseases (e.g., aspiration, trauma, asthma, etc.)
The major clinical problems lie (a) in distinguishing between infection and infarction and (b) in establishing the clinical significance of fat embolism.


The diagnosis of pneumonia is established by culture of a respiratory pathogen from sputum or blood. For many years the most common pathogen has been assumed to be S. pneumoniae based on the seminal studies of Barrett-Connor 25 years ago(4,5). More recent data indicate that those studies are no longer relevant in 1995. Numerous epidemiological studies in normal people indicate a changing epidemiology for pneumonia in recent years. For example, studies(6) in armed forces personnel indicate that the attack rate of pneumonia in the 1990's (77.6/100,000/yr) is one-fourth that of the 1970's (307.6/100,000/yr.). Furthermore, the frequency of S. pneumoniae has declined substantially while that of Mycoplasma pneumoniae has increased so that the two are nearly equal in frequency. Thus the epidemiology of pneumonia in the general population has been changing. A definite etiology is not established by culture in 65 to 75 percent of cases of pneumonia.

Recent studies in SCD show similar findings. Poncz et al. (7) studied 102 episodes of ACS in 70 patients, using careful culture techniques, and found that they could document bacterial infection in only 12 episodes. The single most common cause was mycoplasma (16%), and viral disease at 8% was almost as frequent as the common bacterial agents. In 66% of the episodes, no etiology was established. Nearly identical findings were seen by Sprinkle et al. (8). The report of fulminant mycoplasma pneumonia in the normal population(9) is worrisome for those with sickle cell disease(10). On the other hand, in Curacao(11), 43% of sputum cultures were positive in 81 episodes seen in 53 patients. The most common bacterium was H. Influenza. Both S. aureus and Klebsiella pneumoniae were more common than S. pneumoniae. Other reports have stressed the evolving importance of Chlamydiae pneumoniae and legionella; even tuberculosis, cryptococcus and cytomegalovirus have been reported in SCD. Viral agents causing ACS include influenza, respiratory syncytial virus, CMV, parvovirus, adenovirus, and para influenza. Prophylactic penicillin, Pneumovax and HiB vaccination may contributed to the changing epidemiology for SCD.

However, all studies stress the fact that the majority of ACS episodes cannot be proved to be of infectious origin, and non-infectious causes must be sought very carefully. Vichinsky et al. (12) recently published a study in which 12 of 27 episodes of ACS had evidence for fat embolism as the cause. This is the first documentation of a high frequency for fat embolism in the ACS. The preliminary result from the Multi-Institutional Study of Hydroxyurea in SCD, indicating a significant reduction in ACS in those on hydroxyurea, suggests that a substantial number of ACS episodes are secondary to vascular obstruction(13). Thus our focus on infection as the pre-eminent problem may have been misplaced. Currently, a cooperative study is attempting to establish the relationships among all of the causes of ACS listed in Table 1.


Sputum and blood cultures should continue to be done--despite the historically low yield--so as to identify bacterial agents when present. Recent data indicate that fiber optic bronchoscopy and broncho-alveolar lavage (BAL) provide higher quality specimens for culture and microscopic examination, thus improving confidence in a negative result(14). The safety of bronchoscopy has been established in critically ill patients such as those with HIV infection. Examination of deep sputum and/or BAL specimens for fat-laden macrophages is useful in the diagnosis of the fat Embolism Syndrome (FES) as is examination for trunkal petechiae and lipemia retinalis(15). Other modalities such as thin section CT scan can demonstrate chronic fibrotic changes but its utility in making the distinction between infection/infarction/embolism, etc., has not been established(16).


Antibiotics are generally given although the studies of Charache(17) and Davies(18) do not indicate that antibiotic treatment shortens the clinical course. In view of the current epidemiology studies, the best choice is single agent treatment with erythromycin, but the decision should be guided by knowledge of local bacterial patterns and by the results of sputum smear analysis. I can only condemn the current tendency to give industrial strength doses of a "gorillacillin" as lacking logic and being more likely to produce superinfection.

Serial chest x-rays are needed to assess the extent and course of the pulmonary changes. However, serial arterial blood gas determinations on an FiO2 of 0.21 provide a clearer picture of ongoing pulmonary function as the chest x-ray often lags physiologic events. Arterial blood gas measurements can be replaced by pulse oximetry, especially if there have been simultaneous determinations in the patient so as to establish the correlation between the two. Oxygen saturation and/or the alveolar-arterial oxygen gradient, rather than oxygen tension, provide more relevant information(19). Careful attention to clinical parameters (vital signs, respiratory effort, overall status) is a key ingredient in assessment.

The mainstay of successful treatment is high quality supportive care. Consultation with pulmonary, infectious disease and intensive care specialists is a necessary part of management. Fluid management, oxygenation, chest physiotherapy, bronchodilators, and intermittent incentive spirometry are essential elements of management in the absence of a specific therapy that consistently ameliorates clinical course.


The role of transfusion support is not clearly defined, although there are sporadic case reports of rapid reversal of chest x-ray findings and symptoms immediately post transfusion. It seems clear that there are at least two broad indications for transfusion. Simple transfusion of one or two units can be given to raise the hemoglobin level whenever there is a need for an increase in oxygen carrying capacity. The clinical objective is to make the patient more comfortable, and the indications include moderately severe anemia, high cardiac output, easy fatigability, etc.

On the other hand, exchange transfusion should be given promptly whenever there is evidence of clinical deterioration. Exchange transfusion has the obvious benefit of replacing sickle cells with normal cells and preventing further sickling induced damage to the lungs or other organs, while providing better oxygen carrying capacity. The indications for exchange transfusion in the ACS have not been fully defined but include poor pulmonary function as evidenced by:

  1. arterial oxygen saturation (SaO2) persistently less than 80% despite aggressive ventilatory support

  2. serial decline in SaO2

  3. unstable &/or worsening vital signs

  4. persistent respiratory rate greater than 30/minute

In addition to these indications, exchange transfusion may have a specific therapeutic action in the FES where normal red cells may bind the free fatty acids, preventing further pulmonary damage. Finally, exchange should be considered whenever there is general clinical evidence of a declining course. The objective of transfusion in these situations is life-saving, since progressive worsening of clinical status is associated with high mortality. Exchange transfusion should be done early in the course so that it has prophylactic benefit rather than late when the patient is clearly moribund.

Transfusion carries three risks that are particularly relevant to the sickle cell patient: transmission of viral disease, allo-immunization and acute hyperviscosity. Evidence for hepatitis B infection has been found in 20% of these patients(20), while the data for hepatitis C indicate that the prevalence may be as high as 60%. Cirrhosis due to chronic active hepatitis B or C may be combined with hemosiderosis in these patients. Allo-immunization data from the pre-operative transfusion study indicate a 10% rate of new allo-antibody formation from a transfusion intervention (Vichinsky E, personal communication, 1995). Subsequent transfusion therapy has an increased risk of delayed hemolytic transfusion reaction; such reactions are extremely difficult to treat since further transfusion is usually ineffective in maintaining the hematocrit. An extended phenotype match can reduce the incidence of allo-immunization(21). The hyperviscosity syndrome may occur with mixtures of A and S cells at near normal hematocrits(22). The syndrome is characterized by hypertension and altered mental status and/or seizure activity. Screening donor units for sickle cell trait and maintaining the post transfusion hematocrit at levels less than 35% may prevent this syndrome.

It is widely believed that a post transfusion level of hemoglobin S less than 10% is therapeutically superior to one of 20% or 30%. While data in support of this contention is lacking, the hypothesis is intuitively attractive. The simplest method of monitoring the post transfusion hemoglobin S level is by determining the percentage of sickled cells in a meta-bisulfite preparation. Alternatively, hemoglobin S can be measured by one of the common column chromatography methods.


  1. Johnson CS, Verdegem TV. Pulmonary complications of sickle cell disease. Seminars in Respiratory Medicine. 9:287-93, 1988.
  2. Castro O, Brambilla DJ, Thorington B, Reindorf CA, Scott RB, Gillette P, Vera JC, Levy PS. The acute chest syndrome in sickle cell disease: incidence and risk factors in the cooperative study of sickle cell disease. Blood. 84:643-9, 1994.
  3. Powars D, Weidman JA, Odom-Maryon T, Niland JC, Johnson CS. Sickle cell chronic lung disease: prior morbidity and the risk of pulmonary failure. Medicine. 67:66-76, 1988.
  4. Barrett-Connor E. Acute pulmonary disease and sickle cell anemia. American Review of Respiratory Diseases. 104:155-65, 1971.
  5. Barrett-Connor E. Pneumonia and pulmonary infarction in sickle cell anemia. Journal of the American Medical Association. 224:997-1000, 1973.
  6. Gray GC, Mitchell BS, Tueller JE, Cross ER, Amundson DE. Pneumonia hospitalizations in the US Navy and Marine Corps: rates and risk factors for 6,522 admissions, 1981-1991. American Jounal of Epidemiology. 139:793-802, 1994.
  7. Poncz M, Kane E, Gill FM. Acute chest syndrome in sickle cell disease: etiology and clinical correlates. Journal of Pediatrics. 107:861-6, 1985.
  8. Sprinkle RH, Cole T, Smith S, Buchanan GR. Acute chest syndrome in children with sickle cell disease: a retrospective analysis of 100 hospitalized cases. American Journal of Pediatriatric Hematology/Oncology. 8:105-110, 1986.
  9. Chan ED, Welsh CH. Fulminant mycoplasma pneumoniae pneumonia [clinical conference]. Western Journal of Medicine. 162:133-142, 1995.
  10. Becton DL, Friedman HS, Kurtzberg J, Chaffee S, Falletta JM, Kinney TR. Severe mycoplasma pneumonia in three sisters with sickle cell disease. Pediatric Hematology Oncology. 3:259-65, 1986.
  11. van Agtmael MA, Cheng JD, Nossent HC. Acute chest syndrome in adult Afro-Caribbean patients with sickle cell disease: analysis of 81 episodes among 53 patients. Archives of Internal Medicine. 154:557-61, 1994.
  12. Vichinsky E, Williams R, Das M, Earles A, Lewis N, Adler A, McQuitty J. Pulmonary fat embolism: a distinct cause of severe acute chest syndrome in sickle cell anemia. Blood. 83:3107-12, 1994.
  13. Charache S, Terrin ML, Moore RD, Dover GJ, Barton FB, Eckert SV, McMahon RP, Bonds DR. Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia. Investigators of the multicenter study of hydroxyurea in sickle cell disease. New England Journal of Medicine. 332:1317-22, 1995.
  14. Kirkpatrick MB, Johnson-Haynes J. Jr., Bass JB, Jr. Results of brochoscopically obtained lower airway cultures from adult sickle cell disease patients with the acute chest syndrome. American Journal of Medicine. 90:206-10, 1991.
  15. Chastre J, Fagon JY, Soler P, Fichelle A, Dombret MC, Huten D, Hance AJ, Gilbert C. Bronchoalveolar lavage for rapid diagnosis of the fat embolism syndrome in trauma patients. Annals of Internal Medicine. 113:583-8,1990.
  16. Bhalla M, Abboud MR, McLoud TC Shepard JA, Munden MM, Jackson SM, Beaty JR, Laver JH. Acute chest syndrome in sickle cell disease: CT evidence of microvascular occlusion. Radiology. 187:45-9, 1993.
  17. Charache S, Scott JC, Charache P. 'Acute chest syndrome' in adults with sickle cell anemia: microbiology, treatment and prevention. Archives of Internal Medicine. 139:67-9,1979.
  18. Davies SC, Luce PJ, Win AA, Riordan JF, Brogovic M. Acute chest syndrome in sickle-cell disease. Lancet. I:36-8, 1984.
  19. Emre U, Miller ST, Rao SP, Rao M. Alveolar-arterial oxygen gradient in acute chest syndrome of sickle cell disease. Journal of Pediatrics. 123:272-5, 1993.
  20. Johnson CS, Omata M, Tong MJ, Simmons JF, Jr., Weiner J, Tatter D. Liver involvement in sickle cell disease. Medicine. 64:349-56, 1985.
  21. Sosler SD, Jilly BJ, Saporito C, Koshy M. A simple, practical model for reducing alloimmunization in patients with sickle cell disease. American Journal of Hematology. 43:103-6, 1993.
  22. 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. 27:228-33, 1987.