Localized Leukemic Pulmonary Infiltrates: DISCUSSION

This report describes three patients who presented with respiratory complaints and roentgenographic findings of a localized pulmonary infiltrate as a com­plication of acute leukemia. Previous reports have suggested that most localized lung infiltrates in pa­tients with leukemia are caused by bacterial infec­tion. All of our patients had signs and symptoms suggestive of an acute bacterial pneumonia; however, in each case, specimens obtained from transbronchial biopsy disclosed leukemic infiltrates and no evidence of infection. Early diagnosis of leukemic pulmonary infiltrates enabled the physicians to discontinue anti­biotic therapy and promptly initiate chemotherapy for leukemia. These three patients appear to be the first reported cases in the English literature of leukemic involvement of the lungs presenting as symptomatic focal infiltrates and the first reported cases diagnosed by transbronchial biopsy specimens.

Autopsy studies have reported leukemic involve­ment in the lungs in a remarkably high proportion (24 to 64 percent) of patients dying of acute leukemia.Leukemic invasion of the lungs must be distinguished from aggregation of leukemic blast cells within the pulmonary vasculature which is frequently present in patients with very high blood leukocyte counts. True leukemic infiltrates are typically interstitial in distri­bution; the blast cells are characteristically found around smaller bronchi and blood vessels and may seem to form nodules. Subpleural infiltrates are oc­casionally observed as well. However, leukemic cells are rarely seen within the alveoli on histologic exami­nation unless alveolar hemorrhage has occurred.
cialis professional online

The autopsy studies have indicated that leukemic involvement of the lung usually causes no respiratory signs, symptoms, or roentgenographic abnormalities. Less than 10 percent of the patients found to have leukemic involvement of the lung at postmortem examination complained of respiratory symptoms be­fore death, and only 8 to 22 percent had abnormal chest roentgenograms. Moreover, in most of these patients, the clinical manifestations could not be attributed to leukemia alone, as simultaneous pulmo­nary infection, hemorrhage, or edema was also found.

Table 1—Leukemic Pulmonary Infiltrates: Clinical Characteristics

WBC

Age,

Type of

Count,/

Percent

Poa,

Roentgenographic

Method of

Resolution

У

Leukemia

Symptoms

cu mm

Blasts

mm Hg

Pattern

Diagnosis

with Therapy

Source

Diffuse

26

AMML

Respiratory distress

ND

ND

ND

Bilateral, interstitial

Autopsy

No therapy

Marsh et al13

64

AMML

Dyspnea, cough

49,000

66

ND

Bilateral, patchy nodules

FOB with BAL

Yes

(chemotherapy)

Rossi et alN

24

AMML

Dyspnea

30,000

ND

ND

Bilateral, interstitial

Open lung biopsy

Yes

(radiation)

Man gal and Grove15

24

AMML

Dyspnea

44,000

97

63

“Pulmonary edema”

Autopsy

No therapy

Armstrong et al16

38

AML

Respiratory distress

72,000

44

35

Bilateral,

reticulonodular

Open lung biopsy

Yes

(chemotherapy)

Prakash et al17

30

AML

Respiratory distress

67,000

51

43

Bilateral, alveolar

Open lung biopsy

Yes

(chemotherapy)

Prakash et al17

23

AMML

Respiratory distress

67,000

51

43

Bilateral, alveolar

Open lung biopsy

Yes

(chemotherapy)

Prakash et al17

39

Undifferen­tiated

Dyspnea, chest pain

147,000

91

ND

Bilateral, nodular

Autopsy

No

(chemotherapy)

Geller18

37

AML

Dyspnea, cough

200,000

96

ND

Bilateral

Autopsy

Yes

(chemotherapy and radiation)

Resnick et alw

45

CML?

Dyspnea

130,000

ND

ND

Bilateral reticulonodular

Autopsy

No therapy

Green et al20

Focal

20

AML

Cough

40,200

95

77t

LLL infiltrate

TBB

Yes

(chemotherapy)

Patient 1

24

ALL

Dyspnea

11,800

53

61

RUL/RML infiltrate

TBB

Yes

(chemotherapy)

Patient 2

68

AML

Dyspnea

23,300

44

74t

RUL nodular infiltrate

TBB

Yes

(chemotherapy)

Patient 3

A computer-assisted search of the English language medical literature identified ten other patients with symptomatic involvement of the lung by acute leuke­mia who were described in detail (Table l). Eight of these ten patients had a myelogenous form of acute leukemia (as did two of our three patients). All ten developed leukemic pulmonary infiltrates at a time when the disease was poorly controlled, either at presentation, shortly after initial administration of therapy, or at relapse. The chest roentgenograms invariably showed diffuse bilateral interstitial or reti- culonodular infiltrates. Overall, the clinical picture resembled that of rapidly progressive opportunistic infection or severe pulmonary edema.

Our three patients are similar to those whose cases were reported by others in that they had uncontrolled acute leukemia at the onset of respiratory symptoms. One of our patients (case 2) had a normal WBC count (11,800/cu mm). Notably, however, in common with the previously reported cases, all three of our patients had a peripheral blast count above 40 percent sug­gesting that a high proportion of blast cells in blood is a regular feature of patients with leukemic pulmonary infiltrates.
canadian antibiotics

Leukemic tissue invasion may correlate better with the absolute number of circulating blasts than the percentage of blasts. In all ten patients we reviewed for whom the data were available, the peripheral blast count per cubic milliliter exceeded 6,000; the average was 49,500.