Serum sickness

Study objectives: Mechanical or inflammatory injury to pulmonary endothelial cells may cause impaired pulmonary gas exchange in acute mountain sickness (AMS) and noncardiogenic pulmonary edema in high-altitude pulmonary edema (HAPE). This study was designed to determine whether markers of endothelial cell activation or injury, plasma E- and P-selectin, were increased after ascent to high altitude, in AMS or in HAPE. Design: We collected clinical data and plasma specimens in control subjects at sea level and after ascent to 4,200 m, and in climbers with AMS or HAPE at 4,200 m. Data analysis was performed using standard nonparamedic statistical methods, and results reported as mean [plus or minus] SD. Setting: National Park Service medical camp at 4,200 m on Mt. McKinley (Denali), Alaska. Patients: Blood samples and clinical data were collected from 17 healthy climbers at sea level and again after ascent to 4,200 m, and from a different group of 13 climbers with AMS and 8 climbers with HAPE at 4,200 m. Climbers with AMS were divided into normoxic (n=7) and hypoxemic (n=6) groups. Measurements and results: Using an enzyme immunoassay technique, plasma E-selectin concentrations were found to be increased in the 17 control subjects after ascent to 4,200 m 17.2 [plus or minus] 8.2 ng/mL) as compared to sea level (12.9:[plus or minus]8.2 ng/ml) (p=0.001). Plasma E-selectin concentrations were also increased in subjects with hypoxemic AMS (30.6 [plus or minus] 13.4 ng(ml) and HAPE (23.3: [plus or minus] 9.1 ng/mL) compared to control subjects at sea level (p=0.009). Increased plasma E-selectin concentration significantly correlated with hypoxemia (p=0.006). Plasma P-selection concentrations were unchanged after ascent to 4,200 m and in subjects with AMS and HAPE. Conclusion: Because E-selectin is produced only by endothelial cells, increased plasma E-selectin after ascent to high altitude and in hypoxemic climbers with AMS and HAPE provides evidence that endothelial cell activation or injury is a component of hypoxic altitude illness. (CHEST 1997; 112:1572-78)

Key words: high-altitude pulmonary edema; inflammation; lung injury; pulmonary edema; selectins

Abbreviations: AMS=acute mountain sickness; BSA=bovine serum albumin; HACE=high-altitude cerebral edema; HAPE=high-altitude pulmonary edema; mAb=monoclonal antibody PBS=phosphate-buffered saline solution; Sa[O.sub.2]=arterial oxygen saturation; TBS=TRIS-buffered saline solution

Impaired pulmonary gas exchange occurs in high-altitude pulmonary edema (HAPE) and in some persons with acute mountain sickness (AMS). HAPE is a form of "noncardiogenic" pulmonary edema that occurs in persons who ascend to altitudes above 2,500 m and remain there for 24 to 48 hours or longer.[1] HAPE is associated with an increased mean pulmonary artery pressure, normal pulmonary capillary wedge pressure,[2] and increased protein and cells in alveolar edema fluid.[3,4] Recognized early, HAPE responds well to supplemental oxygen or descent to a lower altitude. Left untreated, HAPE may rapidly progress to death. AMS is a symptom complex commonly seen a few hours to a few days after ascent to altitudes above 2,500 m and is characterized by headache, anorexia, nausea, insomnia, and malaise.[5] AMS is primarily a neurologic syndrome, but pulmonary function may also occur as indicated by hypoxemia or an increased alveolar-to-arterial oxygen pressure difference.[6,7] It iS not known whether AMS is a precursor to HAPE or if AMS is a separate pathophysiologic process. It is well accepted, however, that AMS is the precursor to high-altitude cerebral edema (HACE), which can occur with HAPE.

The etiology of increased pulmonary vascular permeability in HAPE, and impaired pulmonary gas exchange in AMS, remains obscure. A similar mechanism may be responsible for both, with interstitial but not alveolar edema causing hypoxemia in AMS, and progression to alveolar edema occurring in HAPE. Hultgren[2] suggested that uneven hypoxic pulmonary vasoconstriction leads to recruitment and overdistention of some parts of the pulmonary vascular bed in HAPE. This could then result in mechanical endothelial cell injury and increased permeability, edema caused by shear stress from increased velocity of blood flow[8] or stress failure of pulmonary capillaries due to increased transmural pressure.[9] Alternatively, or concomitantly, an inflammatory reaction triggered by hypoxia or mechanical injury to endothelial cells may cause increased permeability edema in HAPE. In BAL fluid of subjects with HAPE, Schoene et al [3,4] found high concentrations of proteins and leukocytes (primarily macrophages), and markers of inflammation including thromboxane [B.sub.2], leukotriene [B.sub.4], and complement component C5a.

Endothelial cell activation occurs in response to physical and humoral stimulation[10] and may occur in HAPE because of mechanical injury to lung vessels or the local generation of inflammatory mediators. E- and P-selectin, which are adhesion molecules expressed on activated endothelial cells, participate in tethering of leukocytes in areas of inflammation. Under some conditions E- and P-selectin are released into plasma in soluble forms.[11,12] Plasma selectins are increased in a number of inflammatory disorders including acute lung injury (P-selectin)[13] and asthma E-selectin),[14] and suggest activation or injury of pulmonary endothelial cells. To test the hypothesis that endothelial cell activation or injury occurs in high-altitude illness, we obtained plasma specimens from persons ill with HAPE and AMS, as well as control subjects at sea level and again at high altitude, and measured soluble forms of E- and P-selectin.

MATERIALS AND METHODS

Setting of the Study

We studied subjects during the May to june climbing seasons in 1994 and 1995 at the National Park Service and Denali Medical Research Project high-altitude medical camp located at 4,200 m on the West Buttress of Mt. McKinley (Denali), in Alaska.[15] The summit of Denali is 6,150 m. The Liverage ambient barometric pressure was 440 mm Hg, and the ambient temperature ranged from -30 [degrees] C (night) to 0 [degrees]C (day). At the medical camp, a heated shelter is used for care of sick or injured climbers and for research. Approval for this study was obtained from the University of Utah Institutional Review Board.

Participants

The 17 control subjects consisted of volunteers from among climbers in Talkeetna, Alaska (200 m, referred to as "sea level" in the text), preparing for an expedition to the West Buttress route of Denali. Those same climbers served as healthy subjects at 4,200 m after a 4- to 10-day ascent from the air landing strip at 2,100 m on the glacier. All plasma specimens from control subjects were obtained during a "rest day" at the 4,200-m camp. The 13 subjects with AMS and 8 subjects with HAPE were climbers who presented to the 4,200-m medical camp for evaluation of altitude illness and subsequently volunteered to enter the study. None of the participants in the study had any history of serious medical illness.

Definition of HAPE

As in previous studies,3,4 and according to published criteria,[16] the diagnosis of HAPE was made clinically according to symptoms and signs in the setting of a recent ascent to high altitude, with no clinical evidence of another cause for hypoxemia. Symptoms included any two of the following: dyspnea at rest, cough, weakness or decreased exercise performance, and chest tightness or congestion. Signs included crackles in at least one lung field on chest auscultation, severe hypoxemia, and resting tachypnea or tachycardia. Arterial oxygen saturation (Sa[O.sub.2]%) was measured with a digital pulse oximeter (Criticare; Waukesha, Wis) while at rest in the sitting position. Severe hypoxemia was defined as an Sa[O.sub.2]% [is less than] 77 ([is greater than] 3 SDs below the mean Sa[O.sub.2]2% for healthy climbers at 4,200 m on Denali).[17] These clinical criteria for diagnosis of HAPE are consistent with clinical findings in HAPE by other investigators.[18,19]

Definition of AMS and HACE

AMS was diagnosed clinically according to the Lake Louise AMS scoring system.[20] A score of >3 on the AMS self-report questionnaire was used to designate AMS. The AMS clinical assessment score was then added to the self-report score and reported as the AMS score. Hypoxemia with AMS ("hypoxemic AMS") was defined as an Sa[O.sub.2]% [is less than] 80 ([is less than] 2 SDs below the mean Sa[O.sub.2]% for healthy climbers at 4,200 m on Denali).[17] HACE was diagnosed clinically according to published criteria of a change in mental status and ataxia in a person with AMS.[16]

Protocol and Handling of Specimens

After informed consent was obtained for participation in the study, subjects were interviewed and a physical examination performed. A venous blood specimen was obtained from a peripheral arm vein with a 21G specimen tube (Vacutainer) needle and a 10-mL yellow stoppered specimen (Vacutainer) tube (1.5 mL of anticoagulant acid citrate dextrose solution; Becton Dickinson Vacutainer System; Franklin Lakes, NJ). Venous blood samples were centrifuged at 1,400 rpm for 10 min within 2 h of collection. The plasma fraction was removed, separated into several aliquots, and frozen at 4,200 m on Denali. At a later time, frozen specimens were transported by air from Denali to Talkeetna, Alaska, where they were placed oil dry ice and shipped to the University of Utah in Salt Lake City, for storage (-70 [degrees]C analyzed. This protocol for storage and transport of specimens has been used in previous studies on Denali.[3,4]

Enzyme-Linked Immunosorbent Assay for Soluble P-selectin

Microtiter plates (Costar; Cambridge, Mass) were coated with monoclonal antibody (mAb) W40[21] (5 [Mu]g/mL, 100 [Mu]L per well) in carbonate coating buffer (10.6 g [NA.sub.2] C[O.sub.3] in 500 mL distilled [H.sub.2]O), adjusted to pH 9.2 with 1N HCl) and stored at 4 [degrees]C overnight. After washing twice with phosphate-buffered saline solution (TBS) (4.091 g NaCl, 1.211 g TRIS-HCl, in 1 L of distilled [H.sub.2]O), nonspecific bind+ng sites were blocked for 2 h with either 5% nonfat dry milk in TBS or 3% bovine serum albumin (BSA) in TBS. After washing three times with TBS/0.1% octoxynol-9 (Triton; Sigma; St. Louis) (500 [Mu]L Triton X-100 in 500 mL TBS), 100 [Mu]L per well of standard (recombinant truncated P-selectin in TBS/1% octoxynol-9)[21] or unknowns (undiluted citrated plasma) was added and incubated for 1.5 h at 37 [degrees] C, 100% humidity. After washing three times with TBS/1% octoxynol-9, biotinylated mAb S12[21] was added (1.0 [Mu]g/ml, 100 [Mu]L per well) and incubated for 1 h at 37 [degrees]C, 100% humidity. After washing three times with TBS/1% octoxynol-9, avidin peroxidase (Sigma A3151, diluted 1:16,000 in TBS) was added (100 [Mu]L per well) and incubated for 1 h at 37 [degrees]C, 100% humidity. After washing once with TBS/1% octoxynol-9 and three times with TBS, ortho-phenylenediamine was added as the peroxidase substrate (100 [Mu]L per well), allowed to develop for 15 to 30 min, and then stopped with IN [H.sub.2],S[O.sub.4] (50 [Mu]L per well). The absorbance at 490 nm was measured on a microtiter reader utilizing software (Softmax; Molecular Devices). The concentration of P-selectin in unknowns was calculated using a standard curve of truncated P-selectin (1.15 to 230 ng/mL). The limit of detection of P-selectin was 10 ng/mL.

Enzyme-Linked Imunosorbent Assay for E-selectin

Microtiter plates were coated with mAb CY1787 to E-selectin (provided by James Paulson, Cytel Corporation) in carbonate coating buffer (5 [Mu]g/ml, 100 [Mu]L per well) and stored at 4 [degrees] C overnight. After washing three times with phosphate-buffered saline solution (PBS)/0.5% polysorbate (Tween; Sigma), 100 [Mu]L per well of standard (recombinant truncated E-selectin in TBS/1% BSA) (provided by Mark Zukowski, Amgen Corporation) or unknowns (undiluted citrated plasma) was added and incubated for 1.5 h at 37'C, 100% humidity. After washing three times with PBS/0.5% polysorbate, biotinylated anti-E-selectin (No. BBA8; R&D Systems; Minneapolis) was added (1.0 [Mu]g/mL in PBS/1% BSA, 100 [Mu]L per well) and incubated for 1 h at 37 [degrees] C, 100% humidity. After washing three times with TBS/0.1% octoxynol, avidin peroxidase (Sigma A3131; diluted 1:10,000 in TBS/1% BSA) was added (100 [Mu]L per well) and inicubated for 1 h at 37 [degrees] C, 100% humidity. After washing four times with TBS/0.1% octoxynol, ortho-phenlenediamine was added (100 [Mu]L per well) and allowed to develop for 15 to 30 min and then stopped with 1 N [H.sub.2] S[O.sub.4] (50 [Mu] L per well) and absorbance at 490 nm measured. The concentration of E-selectin in unknowns was calculated using a standard curve of recombinant E-selectin (1.15 to 230 ng/mL). The limit of detection of E-selectin was 2.3 ng/mL.

Data Analysis

All comparisons were made using standard nonparametric statistical methods[22] with statistical software (Statview 4.5; Abacus Concepts; Berkeley, Calif) used for calculations. Paired comparison of E- and P-selectin concentrations in the control group at sea level and after ascent to 4,200 m was made with a Wilcoxon signed-rank test. Comparisons of concentrations of plasma E- and P-selectin among all groups was made using a Kruskal-Wallis statistic with a Dunn's test for post hoc comparison to sea level controls. Correlation of Sa[O.sub.2]% with plasma E-selectin concentration was made using a Spearman rank correlation coefficient. A significance level of 95% was used for all statistical comparisons and all values are reported in the text as mean [plus or minus] SD.

RESULTS

Fifteen men and two women with a mean age of 37 years (range, 20 to 58 years) volunteered to enter the study as control subjects. In control subjects, the mean AMS symptom score at 4,200 m was 1.3 [plus or minus] 1.2 and the mean Sa[O.sub.2]% was 86 [plus or minus] 3%. Mean concentrations of plasma E- and P-selectin for the control group at sea level and at 4,200 m are listed in Table 1, and individual data are shown in Figures 1 and 2. Plasma E-selectin concentration significantly increased after ascent to 4,200 m (17.2 [plus or minus] 8.2 ng/mL) as compared to sea level (12.9 [plus or minus] 8.2 ng/mL) (p=0.001). Plasma P-selectin concentration was unchanged (Table 1).

Climbers with AMS were divided into normoxic (n = 7) and hypoxic (n = 6) groups. For the normoxic subjects with AMS, the mean symptom score was 6.1 [plus or minus] 2.0, mean Sa[O.sub.2% Was 84 [plus or minus] 2%, mean age was 39 years (range, 24 to 54 years), and all subjects were men. The concentrations of plasma E- and P-selectin was not different from those of sea level control subjects. Mean values for the group are listed in Table 1, and individual data for plasma E-selectin concentrations are shown in Figure 2.

In the hypoxic AMS group (n=6), the mean AMS symptom score was 8.3 [plus or minus] 3.4, mean Sa[O.sub.2]% was 72 [plus or minus] 5%, mean age was 34 years (range, 25 to 42 years), and one subject was female. In hypoxic climbers with AMS, plasma E-selectin concentration was significantly increased (30.6 [plus or minus] 13.4 ng/ml) as compared to sea level control subjects (12.9 [plus or minus] 8.2 ng/mL) (p [is less than] 0.0001), and plasma P-selectin was unchanged. Mean values are listed in Table 1, and individual data are shown in Table 2 and Figure 2.

In the HAPE group (n=8), the mean age was 35 years (range, 28 to 41 years), all were male, and none smoked cigarettes. Two of eight climbers with HAPE also had HACE. Clinical data on each HAPE subject are fisted in Table 2, and mean values for the group are shown in Table 1. In samples from climbers with HAPE, plasma E-selectin concentration (23.3 [plus or minus] 9.1 ng/mL) was significantly greater than in sea level control subjects (p=0.0078) (Fig 2), and plasma P-selectin was unchanged (Table 1). Plasma E-selectin concentrations were measured after resolution of HAPE in two subjects (subjects 1 and 6) and before onset of HAPE in one subject (subject 8), and each value was lower than the corresponding plasma E-selectin concentration measured during acute presentation with HAPE (Table 2).

Discussion

We found mildly increased plasma concentrations of E-selectin in healthy climbers after ascent to high altitude, and increases of greater magnitude in climbers ill with AMS who were hypoxemic and in climbers ill with HAPE as compared to healthy sea level control subjects (Table 1, Figs 1-2). In a preliminary study conducted at 4,200 m on Denali in cooperation with our laboratory, Elridge and colleagues[23] also reported an increase in plasma E-selectin in seven healthy climbers after ascent to 4,200 m, and a nonsignificant increase in plasma E-selectin in four subjects with HAPE; subjects with hypoxic AMS were not studied. Because stimulated endothelial cells are the only known source of soluble E-selectin in the circulation,[11,12,24] these observations suggest that endothelial cell activation is a component of high-altitude illness.

E-selectin is synthesized solely by endothelial cells. Neither E-selectin messenger RNA nor protein is found in resting, unactivated endothelial cells in vitro; however, its expression is induced within hours of activation by inflammatory cytokines, including interleukin-1 and tumor necrosis factor-[Alpha], or by lipopolysaccharide and other endothelial toxins.[11,12] In vivo, a variety of inflammatory conditions are associated with E-selectin expression by endothelium.[14,24,25] although direct induction of E-selectin by hypobaric hypoxia or mechanical stresses has not been shown, hypoxia enhances the synthesis of E-selectin in bovine aortic endothelial cells stimulated with bacterial lipopolysaccharide or tumor necrosis factor-[Alpha].[26] It is unknown, however, if hypoxia potentiates E-selectin expression when human endothelial cells are stimulated with cytokines or other mediators. Our results suggest an association between exposure to hypobaric hypoxia and E-selectin expression in humans. In addition to increased plasma E-selectin concentration after ascent to high altitude in control subjects, and increased plasma E-selectin concentration in hypoxic subjects with AMS and HAPE, we found a significant inverse correlation between Sa[O.sub.2]%and plasma E-selectin concentration at high altitude (p=0.006).

E-selectin is a transmembrane glycoprotein that is transiently expressed for hours on the surfaces of stimulated endothelial cells, where it mediates the tethering of leukocytes, and then is reinternalized.[10-12] In vitro, a soluble form of E-selectin is also released from stimulated endothelial cells[27,28] by a mechanism that may involve proteolytic cleavage.[24,27] While these observations provide a basis for the presence of a circulating, soluble form of E-selectin in the plasma of normal subjects, and for its detection in plasma samples from subjects with AMS and HAPE, the specific vascular bed(s) from which it is derived and the stimuli driving its production in normal subjects are unknown. The fate of E-selection after it is released into the plasma is also unknown. Thus, we cannot completely exclude the possibility that the elevated levels of E-selectin that we found in subjects with HAPE and hypoxic AMS are due to decreased degradation, rather than increased production and release. However, increased circulating E-selectin concentrations have been reported in a variety of syndromes of inflammation and vascular injury,[14,24,25] suggesting that the increases in subjects with hypoxemic altitude illness in our study are due to endothelial cell activation.

In contrast to the results with E-selectin, we found that the levels of P-selectin in plasma samples from climbers with hypoxic altitude illness were not increased compared to samples from control subjects. The regulation of P-selectin in endothelial cells is considerably different from that of E-selectin.[10-12] P-selectin is constitutively present in Weibel-Palade bodies of resting human endothelial cells and is rapidly translocated to the cell surface in response to agonists different from those that induce the synthesis of E-selectin. Membranous P-selectin is then reinternalized by endothelial cells after a variable period on the plasma membrane that is generally much shorter than that of E-selectin.[10-12] Endothelial cells contain a soluble form of P-selectin that lacks a transmembrane domain and is coded for by an alternatively spliced messenger RNA; the absence of a hydrophobic transmembrane domain predicts that this variant of P-selectin will be secreted.[12,21] Using an assay different from the one that we employed, Ushiyama et al[21] reported soluble P-selectin in plasma from normal subjects, with values very similar to those we found in plasma from healthy climbers at sea level. Others have also reported that P-selectin is present in the blood of normal subjects.[29,30] In addition, increased levels of P-selectin have been found in plasma samples from patients with vascular injury and thrombosis (our unpublished observations).[13,29,30] The factors that influence the secretion and/or release of the soluble form of P-selectin under these conditions are unknown.

The fact that we did not find elevated levels of soluble P-selectin in the blood of subjects with hypoxemic altitude illness, whereas E-selectin was elevated (Fig 2), may be due to differences in the spectrum of inflammatory agonists that induce expression and/or release of the two molecules.[10-12] The finding suggests that local generation of thrombin -- which is an agonist for expression of P-selectin but not E-selectin -- is not a major feature of HAPE. Thrombin might be expected to be generated if there is "stress failure" of pulmonary endothelium and exposure of thrombogenic subendothelial matrix occurs,[9] or if procoagulant stimuli[19] are generated in this condition. We cannot exclude the possibility that there is expression of P-selectin on endothelial plasma membranes without release of the soluble form in subjects with hypoxic AMS or HAPE. Our findings, however, are consistent with those of other investigators who found no increase in P-selectin in lung lavage fluid of climbers ill with HAPE.[31]

Our observation that the level of plasma E-selectin is elevated in subjects with AMS associated math hypoxia, but not normoxic subjects with AMS (Fig 1), suggests that pulmonary endothelium is a source of the circulating protein, although we cannot exclude other vascular beds. The higher plasma E-selectin concentration in some subjects with AMS may be due to endothelial cell activation in vascular beds other than the lung, as peripheral edema, mild cerebral edema, and proteinuria are known to occur with AMS and may be due to vascular injury in the brain, kidney, and other tissues.[5] The finding of increased plasma E-selectin in subjects who are hypoxemic with AMS and in subjects with HAPE, however, is also consistent with vascular injury or inflammation in the lung and with the possibility that AMS with hypoxemia, as we clinically define it, is a precursor to HAPE. We recognize that the diagnoses of AMS with hypoxemia and HAPE in our study relied on clinical criteria that may not have precisely distinguished the two conditions; subjects with AMS who were hypoxemic may have had early HAPE with interstitial, but not alveolar, lung edema.

Our observations suggest that expression of E-selectin by endothelial cells in hypoxemic altitude illness, possibly triggered by inflammatory mediators in the setting of severe hypoxia, may amplify inflammation in the lung leading to increased permeability edema. We do not doubt that high pulmonary vascular pressures play an important role in HAPE, but our findings, and those of others who found increased inflammatory mediators in BAL fluid of subjects math HAPE,[3,4,31] suggest that inflammation is also a component of this syndrome. Clearly the inflammatory response in HAPE is not as severe as in other forms of lung injury, but may still result in significant increased permeability edema in the setting of high pulmonary vascular pressure. Mild inflammation would also be more easily reversible once pulmonary vascular pressure is lowered, as occurs with oxygen administration or descent in persons ill with HAPE. Further analysis of endothelial selectins, and of their relationship to other indexes of endothelial injury and activation in hypoxemic altitude illness, may provide important insights into the molecular variables that influence lung edema in high-altitude illness and other forms of acute lung injury. They will also be useful in further characterization of the biology of endothelial selectins in human disease.

REFERENCES

[1] Schoene RB. Pulmonary edema at high altitude: review, pathophysiology, and update. Clin Chest Med 1985; 6:491-507

[2] Hultgren HN. High altitude pulmonary edema. In: Staub N, ed. Lung water and solute exchange. New York: Marcel Dekker, 1978: 437-69

[3] Schoene RB, Swenson ER, Pizzo CJ, et al. The lung at high altitude: bronchoalveolar lavage in acute mountain sickness and pulmonary edema. J Appl Physiol 1988; 64:2605-13

[4] Schoene RB, Hackett PH, Henderson WR, et al. High altitude pulmonary edema: characteristics of lung lavage fluid. JAMA 1986; 256:63-69

[5] Selioene RB, Hackett PH, Hornbein TF. High altitude. In: Murray TF, Nadel Ja, eds. Textbook of respiratory medicine. Philadelphia: WB Saunders, 1994; 2062-98

[6] Grissom CK, Roach RC, Sarnquist FH, et al. Acetazolamide in the treatment of acute mountain sickness: clinical efficacy and effect on gas exchange. Ann Intern Med 1992; 116: 461-65

[7] Sutton JR, Bryan AC, Gray GW, et al. Pulmonary gas exchange in acute mountain sickness. Aviat Space Environ Med 1976; 47:1032-37

[8] Staub NC. Pulmonary edema due to increased microvascular permeability to fluid and protein. Circ Res 1978; 43:143-51

[9] Tsukimoto K, Mathieu-Colstello O, Prediletto R, et al. Ultrastructural appearances of pulmonary capillaries at high transmural pressures. J Appl Physiol 1991; 71:573-82

[10] Zimmerman GA, Prescott SM, McIntyre TM. Endothelial cell interactions with granulocytes: tethering and signaling molecules. Immunol Today 1992; 13:93-100

[11] Bevilacqua MP, Nelson RM. Selectins. J Clin Invest 1993; 91:379-87

[12] McEver RP. Selectins. Curr Opin Immunol 1994; 6:75-84

[13] Sakamaki F, Ishizaka A, Handa M, et al. Soluble form of P-selectin in plasma is elevated in acute lung injury. Am J Respir Crit Care Med 1995; 151:1821-26

[14] Montefort S, Lai CKW, Kapahi P, et al. Circulating adhesion molecules in asthma. Am J Respir Crit Care Med 1994; 149:1149-52

[15] Hackett PH, Roach RC, Schoene RB, et al. The Denali medical research project, 1982-1985. Am Alpine j 1986; 28:129-37

[16] The Lake Louise consensus on the definition and quantification of altitude illness. In: Sutton JR, Coates G, Houston CS, eds. Hypoxia and mountain medicine. Burlington, VT: Queen City Printers, 1992; 327-30

[17] Hackett PH, Roach RC, Hollingshead F, et al. Arterial saturation predicts subsequent acute mountain sickness. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and cold. New York: Praeger, 1987; 544

[18] Vock P, Christian F, Franciolli, et al. High-altitude pulmonary edema: findings at high-altitude chest radiography and physical examination. Radiology 1989; 170:661-66

[19] Bartsch P, Haeberli A, Franciolli M, et al. Coagulation and fibrinolysis in acute mountain sickness and beginning pulmonary edema. J Appl Physiol 1989; 66:2136-44

[20] Roach RC, Bartsch P, Hackett PH, et al. The Lake Louise acute mountain sickness scoring system. In: Sutton JR, Coates G, Houston CS, eds. Hypoxia and molecular medicine. Burlington, Vt: Queen City Printers, 1993; 272--t4

[21] Ushiyama S, Laue TM, Moore KL, et al. Structural and functional characterization of monomeric soluble P-selectin and comparison with membrane P-selectin. J Biol Chem 1993; 268:15299-37

[22] Glantz SA. Primer of bio-statistics. 3rd ed. New York: McGraw-Hill, 1992

[23] Eldridge MW, Johnson DH, Hill HR. Cytokines and adhesion molecules in high altitude pulinonary edema. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and the brain. Burlington, Vt: Queen City Printers, 1995; 243-50

[24] Gearing AJH, Newman W. Circulating adhesion molecules in disease. Immunol Today 1993; 14:506-12

[25] Cowley AJH, Heney D, Gearing AJH, et al. Increased circulating adhesion molecule concentrations in patients with the systemic inflammatory response syndrome: a prospective cohort study. Crit Care Med 1994; 22:651-57

[26] Zund G, Nelson DP, Neufeld EJ, et il. Hypoxia enhances stimulus-dependent induction of E-selectin on aortic endothelial cells. Proc Natl Acad Sci USA 1996; 93:7075-80

[27] Newman W, Beall LD, Carson CW, et al. Soluble E-selectin is found in supernatants of activated endothelial cells and is elevated in the serum of patients with septic shock. J Immunol 1993; 150:644-54

[28] Leeuwenberg JFM, Smeets EF, Neefies JJ, et al. E-selectin and intercellular adhesion molecule-1 are released by activated human endothelial cells in vitro. Imminulogy, 1992; 77:543-49

[29] Ishiwata N. Takio K, Katayama M, et, al. Alternatively spliced isoform of P-selectin is present in vivo as a soluble molecule. J Biol Chem 1994; 269:23708-15

[30] Katayama M, Handa M, Araki Y, et al. Soluble P-selectin is present in normal circulation and its plasma level is elevated in patients with thrombotic thrombocytopenic purpura and haemolytic uraemic syndrome. Br J Haematol 1993; 84: 702-10

[31] Kubo K, Hanaoka M, Yamaguchi S, et al, Cytokines in bronchoalveolar lavage fluid in patients with high altitude pulmonary oedema at moderate altitude in japan. Thorax 1996; 51:,-39-42

COPYRIGHT 1997 American College of Chest Physicians
COPYRIGHT 2004 Gale Group