Diagnosis of cellulitis in the immunocompromised host

A prospective study of diagnostic techniques in cellulitis was performed on 28 patients with malignancy. Twenty-two (78%) of the fine needle aspiration cultures and 10 (35%) of the blood cultures were positive in this immunocompromised population. The incidence of positive fine needle aspiration cultures (P<0.005) or bacteremia (P<0.0005) was significantly higher than results obtained in an immunocompetent population with cellulitis at the same institution. Staphylococci or streptococci were recovered in 59% of positive cultures, while aerobic Gram-negative bacilli grew in 33%. This study indicates that in the immunocompromised population with cellulitis, fine needle aspiration and blood cultures should be obtained, and the antibiotic regimen should cover Gram-positive cocci and Gram-negative bacilli pending the results of cultures.


PATIENTS AND METHODS
Patients with a previous diagnosis of malignancy admitted to Truman Medical Center over a period of 28 months with a diagnosis of acute cellulitis were included in the study after informed consent was obtained.Cellulitis was diagnosed by the clinical find ings described above.Patients were excluded from the study if the cellulitis was associated with ulcerative lesions , abscesses, underlying osteomyelitis, or concurrent antibiotic therapy.
Fine needle aspiration of the leading edge of the cellulitis was performed according to the procedure described by Uman and Kunin in 1974 (5).The skin was disinfected with povidone-iodine and alcohol, and a sterile 21 or 22 gauge needle inserted without local anesthetic at the leading edge  of the infection and aspirated.If no material was recovered in the syringe, 1.0 mL of nonbacteriostatic saline solution was injected into the subcutaneous tissue and aspirated.The recovered material was inoculated onto the following m edia: tryptic soy blood agar, colistin-nalidixic acid agar, MacConkey agar, chocolate agar, and a liquid thioglycolate medium.In addition, a erobic and anaerobic blood cultures were obtained.
Experimental data were analyzed by Student's t test.

RESULTS
1Wen ty-eight patients were enrolled in the s tudy.There were 13 male and 15 female p atients between the ages of 18 and 82 years with a mean age of 46 .The sites of infection included : upper extremity 18 (64%); lower extremity eigh t (28%); and one each (3.5%) of breast (lung cancer) and perineum (ovarian cancer).Eight of the 28 patients had neutrophil counts less than 1000/mm 2 .1Wenty-two (78%) of the fine needle aspiration cultures were positive.Blood cultures were positive in 10 patients (35%), including six of eight n eutropenic patients (Table 1) .The organisms isolated in the blood cultures were identical to the bacteria recovered from fine needle aspiration cultures in all cases (kappa=l).Staphylococcus aureus or group A streptococci were recovered in 13 cultures (59%) (

DISCUSSION
This study focused on cellulitis in the immunocompromised host.The high rate of positive cultures in pa tients with impaired immunity is probably secondary to increased numbers of infectious pathogens at the site of infection.Immunosuppression in patients with malignancies is secondary to effects of the neoplasm itself and the treatmen t modalities used in oncology.These defects consist of a decreased number of granulocytes, qualitative abnormalities in n eutrophils, abnormal immunoglobulins, impaired cellmediated immunity, and an impaired skin barrier (1 ,6 ,7).
The results also showed a high ra te of bacter emia with the causative organism of the cellulitis.Others have reported positive blood culture ra tes in cellulitis of from 0 to 16% (8,9) .Interestingly, bacteremia in the present study was remarkably high (38.5%)even compared to patients with cellulitis with granulocyte counts less than 100/mm 3 secondary to malignancy (19%) (10).
The value of fine needle aspiration is still debated , with sensitivities ranging from 5 to 64% in patients with a wide spectrum of underlying diseases (Table 3) (8,11) .Epperly (12), who performed the only study exclusively on patients without underlying disease, reported nine of 103 patients (8 .7%) with cellulitis having positive aspiration cultures, which all grew staphylococcal or streptococcal species.Kielhofner's study (4) pointed out the value of fin e needle aspiration in immunocompromised hosts, especially diabetics .In a comparison of pa tient s with cellulitis with no underlying disease a t the a uthors' institution u s ing the same methods , there was a significant increas e in positive fine needle aspiration cultures (P<0.005) and ba ct eremia (P<0.0005) in patients with malignancies, and an increase incidence of aerobic Gram-negative infection (P<0.05).
The present results reaffirm fine needle aspiration as a sensitive and safe method for determining the etiology of cellulitis in this select patient population.Because of the higher incidence of aerobic Gram-negative bacilli and a significant tendency toward bacteremia, the authors recommend initial antibiotic coverage for both Gramnegative bacilli and Gram-positive cocci pending the results of cultures.

TABLE 1
Sensitivity of culture techniques for cellulitis in immunecompromised patients Underlying disease Needle aspirate culture Blood culture Number Positive (%) Positive (%) Lymphoreticular

TABLE 3
Comparison of acute cellulitis studies in adults Study Ref. Fine needle aspirate Blood culture results