|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Surgery I, National Defense Medical College, 3-2, Namiki, Tokorozawa, Saitama, 359, Japan.
Correspondence to: Manabu Kinoshita, MD. Department of Surgery, Gifu Military Hospital, Gifu Air Base, Naka, Kakamihara, Gifu, 504-8701, Japan; e-mail: ishi2{at}ylw.mmt.or.jp
| Abstract |
|---|
|
|
|---|
Design: Retrospective study.
Setting: Surgical unit in a tertiary-care university hospital.
Patients: Thirty-three patients who died of intra-abdominal sepsis and received an autopsy.
Measurements: Just before each patients death, (1) plasma endotoxin was determined by the limulus gelation test, and (2) the severity of ALI was estimated by the Murray lung injury score. (3) Neutrophil accumulation in the pulmonary microcirculation was evaluated in the autopsy specimens using a computerized picture analysis method.
Results: (1) Endotoxin-positive patients were more likely to fall into the severe lung injury group (endotoxin-positive patients, 38% vs endotoxin-negative patients, 0%; p < 0.01). (2) The endotoxin-positive patients exhibited significantly higher neutrophil accumulation in the pulmonary microcirculation than endotoxin-negative patients (8,349 ± 984/mm2 vs 4,047 ± 447/mm2, respectively; p < 0.01). (3) Severe lung injury patients with endotoxemia had almost the same degree of neutrophil accumulation in the pulmonary microcirculation as mild-to-moderate lung injury patients with endotoxemia (8,338 ± 1,622/mm2 vs 8,359 ± 1,290/mm2, respectively), showing a significant higher neutrophil accumulation compared to no lung injury patients without endotoxemia (5,102 ± 410/mm2; p < 0.01).
Conclusion: Endotoxemia might cause ALI and pulmonary neutrophil accumulation. Pulmonary neutrophil accumulation might not be enough to cause severe lung injury (ARDS), although it is necessary to cause ALI, because the degree of pulmonary neutrophil accumulation did not correlate with the severity of ALI.
Key Words: acute lung injury ARDS autopsy study endotoxemia neutrophil accumulation
| Introduction |
|---|
|
|
|---|
Several reports have revealed a predominant increase of neutrophils in BAL fluid (BALF) obtained from patients with the most severe form of ALI, ARDS.7 8 Several pathologic studies also have demonstrated prominent neutrophil sequestration and migration into interstitial and alveolar spaces of the lung at autopsy in patients with ARDS.9 10 11 Results of these human studies do not conflict with those of experimental studies using animal models. Nevertheless, there are few precise quantitative studies that especially focus on pulmonary neutrophil accumulation in patients with endotoxemia. In this clinical study, we investigated quantitatively neutrophil accumulation in the pulmonary microcirculation in patients with ALI following endotoxemia, to elucidate the role of neutrophils in the pathogenesis of ALI following endotoxemia.
| Materials and Methods |
|---|
|
|
|---|
|
Methods
Endotoxin Determination:
Plasma endotoxin was determined in
all registered patients by the limulus lysate gelation
method.12
In 31 patients, blood samples for endotoxin
determination were obtained during the agonal period at the same time
as both the determination of the peripheral neutrophil count and the
evaluation of ALI. The agonal period meant a time within a few hours
before death. In the other two patients, whose infectious conditions
had been quite stable until their deaths, the determination of
endotoxin was performed using samples obtained 4 or 5 days before
death. Samples were collected in sterile, pyrogen-free glass tubes
containing 3.8% sodium citrate (1.3 x 105 mol/L
blood). They were then centrifuged at 1,000g for
40 s to obtain platelet-rich plasma and stored at - 80°C until
assayed.
Proteins were first removed from stored specimens by adding one-fourth volume chloroform and vigorously mixed for 1 h at room temperature (25°C). Emulsions of chloroform and plasma were separated into three layers by centrifugation at 500g for 10 min. The middle cloudy layer was then extracted for endotoxin determination.
Endotoxin was detected by mixing 0.1 mL of limulus amebocyte lysate (Pregel-S; Seikagaku Kogyo; Tokyo, Japan) and 0.1 mL of the deproteinized material and incubating for 4 h at 37°C. Development of a marked increase in gelation at shaking of a sample was interpreted as evidence for the presence of endotoxin. Mixtures with unchanged gelation following incubation were interpreted as negative. False-positive reactions were eliminated by excluding patients with chronic hepatitis, cirrhosis, or fungal infection.
Evaluation of ALI: The severity of ALI was objectively evaluated during the agonal period at the same time as the determination of peripheral neutrophil count, using the Murray lung injury score.13 This score was obtained by grading the following clinical parameters on a scale of 0 to 4: the extent of chest infiltration on radiographs, the ratio of arterial oxygen tension to inspired oxygen concentration, the positive end-expiratory pressure, and pulmonary compliance. The overall lung injury score was derived by dividing the total score by the number of criteria evaluated. "No lung injury" was defined as a score 0, "mild-to-moderate lung injury" as 0.1 to 2.5, and "severe lung injury" as > 2.5.
The patients were classified into the following three groups according to the severity of lung injury during the agonal period: "no," "mild-to-moderate," and "severe." Patients with underlying pulmonary disease and with any signs of upper respiratory infections on physical examination, chest radiograph, or bacteriologic examination of sputum were excluded from this study. Patients with lung injury of > 1-week duration were also excluded because we focused only on ALI in this study.
Evaluation of Neutrophil Accumulation in the Pulmonary Microcirculation: At autopsy, lung specimens were fixed by injecting 10% formalin into the primary bronchi at a pressure of 100 cm and then immersing in 10% formalin for about 5 days. Microscopic slides were prepared from formalin-fixed, paraffin-embedded sections of the pulmonary tissue and stained with hematoxylin and eosin.
Ten sections of the left upper pulmonary lobe of each patient showing the most typical histopathologic changes were photographed using 35-mm film (X100). Each picture was scanned by the Nikon Coolscan Control 1.2-J scanner (Nikon; Tokyo, Japan) and obtained as a file in Adobe Photoshop 3.0-J software (Adobe Systems; Mountain View, CA). Then, each file was transferred to Aldus Persuasion 3.0-J software (Adobe Systems) to estimate pulmonary neutrophil accumulation precisely. Neutrophils in the capillary and postcapillary vessels, defined as the pulmonary microcirculation, were counted on each image of the pulmonary tissue (Fig 1 , top). The area of the pulmonary microcirculation was estimated as follows: all vessel lumens of the image on Aldus Persuasion 3.0-J were traced, and the total cross-sectional traced areas were then calculated using the TOSPIX picture analysis package (Toshiba; Tokyo, Japan; Fig 1 , bottom). Neutrophil accumulation in the pulmonary microcirculation was quantified in each image by dividing the neutrophil count by the vascular area, and the average value of 10 fields was calculated for each patient.
|
Statistics:
Data are expressed as mean ± SEM. One-way
analysis of variance was used to evaluate group differences in
neutrophil counts. The
2 test was used to demonstrate a
relationship between endotoxemia and the severity of ALI. The
correlation between neutrophil count in the pulmonary microcirculation
and circulating neutrophil count was evaluated by Pearsons
coefficient. For all statistical tests, a p value < 0.05 was
considered significant.
| Results |
|---|
|
|
|---|
Relationship Between Endotoxemia and Severity of ALI
Eight of the 21 patients in the endotoxin-positive group exhibited
severe lung injury, 13 exhibited mild-to-moderate lung injury, and none
exhibited no lung injury. None of the 12 patients in the
endotoxin-negative group had severe lung injury, 7 had mild-to moderate
lung injury, and 5 had no lung injury. The endotoxin-positive group was
significantly more likely to have severe lung injury than the
endotoxin-negative group (p < 0.01; Table 2
).
|
|
|
Correlation Between Neutrophil Count in the Pulmonary
Microcirculation and Circulating Neutrophil Count
The neutrophil count in the pulmonary microcirculation did not
correlate with the circulating neutrophil count (Fig 4
).
|
|
| Discussion |
|---|
|
|
|---|
It is well known that there is a broad range of sensitivity to endotoxin among mammals and organs. For example, endotoxin can cause circulatory shock in anesthetized dogs at a dose that does not produce severe lung injury.14 In unanesthetized sheep, a similar dose of endotoxin that is too small to cause shock produces severe lung injury.15 The systemic response of humans to endotoxin appears to be much more sensitive than these animals. Endotoxin usually has been given to human volunteers at a dose of 2 to 4 ng/kg or 20 U/kg as an IV bolus injection. These doses of endotoxin usually provoke mild but variable responses characteristic of Gram-negative sepsis in human.2 16 17 18 In animal studies, endotoxin is usually given in quantities at least 100 times those given to human volunteers.14 15 Therefore, the findings derived from experimental models of endotoxemia in animals should be cautiously applied to human endotoxemia. Ideally, studies to elucidate the role of neutrophils in ALI following endotoxemia should be performed in human subjects.
Powe et al19 has reported the pulmonary accumulation of neutrophils in four ARDS patients by using a radionuclide scanning technique. This study clarified neutrophil accumulation clinically but did not show any relation to endotoxemia. Our study provided direct evidence of neutrophil pulmonary accumulation in endotoxemia. Since the patients in our study had no underlying pulmonary disease, ALI was considered to be derived from abdominal sepsis. So these patients were suitable for studying the influences of endotoxin on the lung. It is impossible to perform open lung biopsy for such patients, so we evaluated pulmonary specimens obtained at autopsy.
Once endotoxin enters the body, it activates neutrophils, macrophages,
endothelia, and complements.20
21
22
Martich et
al16
and Kuhns et al17
have reported that
injections of endotoxin to volunteers could provoke increases of
interleukin 8 (IL-8) and other cytokines, such as tumor necrosis
factor-
and interleukin-1ß. Donnelly et al23
have
reported high IL-8 concentrations in the BALF of ARDS patients, as well
as IL-8 expression by alveolar macrophages using immunohistochemistry.
It also has been reported that human endothelia release IL-8 in
vitro.24
25
Pulmonary tissue contains tremendous
numbers of alveolar macrophages and capillary vessels that can produce
IL-8. So in human endotoxemia, the lung may be exposed to high
concentrations of IL-8 released from those cells, causing abundant
neutrophil migration to the lung. Endotoxin and evoked cytokines also
could induce adhesion molecules including endothelial leukocyte
adhesion molecule-1, intercellular adhesion molecule-1 on the surface
of endothelia, and macrophage-1 and lymphocyte
function-associated antigen-1 on neutrophils.26
27
28
29
Schleimer et al30
and Read et al31
have
reported that endotoxin and cytokines promote neutrophil adherence on
endothelium in vitro. In human endotoxemia, neutrophils
migrating to the lung may adhere to pulmonary capillary vessels under
these adhesion molecules rich condition.
In the present study, neutrophil accumulation in the pulmonary
microcirculation was significantly higher in the patients with
endotoxemia, whereas the circulating neutrophil count did not
significantly increase. Moreover, there was no correlation between
these two. In some of the registered patients, we determined neutrophil
elastase, which was a complex with
-1 protease inhibitor, or
superoxide anion derived from circulating neutrophil. Those factors did
not correlate with the circulating neutrophil count but tended to
correlate with the degree of neutrophil accumulation in the lung (data
were not shown). We think that the most important factor for neutrophil
accumulation in the lung is not the number of circulating neutrophil
but their activated potential. Some patients complicated bone marrow
suppression due to severe sepsis. It might be also one of the reasons
for the no correlation between the circulating neutrophil count and the
degree of neutrophil accumulation in the lung.
Suffredini et al2 have pointed out that alveolar neutrophils obtained from BALF of volunteers injected with endotoxin did not increase in spite of the emergence of mild pulmonary dysfunction. In the present study, we also found few neutrophils in the alveolar spaces in severe lung injury, even though many neutrophils accumulated in the pulmonary microcirculation. These findings suggest that most of the neutrophils migrating to the lung remain in the pulmonary microcirculation and do not enter the alveolar spaces. The increase of neutrophils in the pulmonary microcirculation appears to be much more important in the pathophysiology of ALI than those in the alveolar spaces.
Although we found significant neutrophil accumulation in the pulmonary microcirculation of the patients with endotoxemia, there was no difference in its degree between the severe and the mild-to-moderate lung injury patients with endotoxemia. The severity of the lung injury did not seem to be related only to the degree of neutrophil accumulation in the pulmonary microcirculation. Not only the accumulation of neutrophils but also neutrophil-derived bioactive substances such as elastase and superoxide anion might be related to deteriorate the lung injury to a more severe condition like ARDS.
Endotoxin can activate complements that are thought to cause ARDS.22 32 33 Several investigators have suggested that complements activated by endotoxin cause the aggregation of neutrophils and leukoemboli in the lung, resulting in ARDS.34 35 However, we found few leukoemboli but many sequestrated neutrophils in the pulmonary microcirculation in patients with ALI. On the other hand, Moore et al18 have reported that neutrophils may be activated by administrating endotoxin to volunteers without concurrent complement activation. Therefore, a neutrophil chemotactic factor IL-8 or neutrophil-endothelium adhesion molecules (endothelial leukocyte adhesion molecule-1, intercellular adhesion molecule-1, macrophage-1, lymphocyte function-associated antigen-1, etc.), rather than leukoemboli, may be primarily responsible for neutrophil pulmonary accumulation seen in ALI following endotoxemia.
In conclusion, significant neutrophil accumulation in the pulmonary microcirculation was found in the patients with endotoxemia, who all were complicated by ALI. Whereas the degree of pulmonary neutrophil accumulation was not related to the circulating neutrophil count nor the severity of ALI, endotoxemia might cause pulmonary neutrophil accumulation in human. Although this accumulation might be necessary to cause ALI, it does not seem to be enough to deteriorate the lung injury to severe lung injury (ARDS).
| Footnotes |
|---|
Received for publication December 15, 1998. Accepted for publication August 10, 1999.
| References |
|---|
|
|
|---|
. J Leukoc Biol 57,820-826[Abstract]
, LPS, IL-1 ß. Science 243,1467-1469
regulates the expression of the adhesion molecule ELAM-1 and IL-6 production by human endothelial cells in vitro. J Immunol 145,2110-2114[Abstract]
This article has been cited by other articles:
![]() |
M. Kinoshita, S. Ono, and H. Mochizuki Neutrophil-Related Inflammatory Mediators in Septic Acute Respiratory Distress Syndrome J Intensive Care Med, November 1, 2002; 17(6): 308 - 316. [Abstract] [PDF] |
||||
![]() |
J. A. Burns, T. B. Issekutz, H. Yagita, and A. C. Issekutz The {beta}2, {{alpha}}4, {{alpha}}5 Integrins and Selectins Mediate Chemotactic Factor and Endotoxin-Enhanced Neutrophil Sequestration in the Lung Am. J. Pathol., May 1, 2001; 158(5): 1809 - 1819. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Raiden, Y. Pereyra, V. Nahmod, C. Alvarez, L. Castello, M. Giordano, and J. Geffner Losartan, a selective inhibitor of subtype AT1 receptors for angiotensin II, inhibits neutrophil recruitment in the lung triggered by fMLP J. Leukoc. Biol., November 1, 2000; 68(5): 700 - 706. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |