| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vol. 84, Issue 3, 782-790, March 1998
Sections of Respiratory Diseases and Critical Care Medicine, Department of Internal Medicine, Department of Pediatrics, Section of Thoracic Surgery, Department of Surgery, and Department of Immunology, University of Manitoba, Winnipeg, Manitoba, Canada R3E OZ3
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
We examined the
effect of anaphylactic shock on the longitudinal distribution of
pulmonary vascular resistance (PVR) in ragweed-sensitized dogs in which
PVR was partitioned into an upstream arterial component (Rus) and a
downstream venous and capillary component (Rds). We also assessed
whether Rus and Rds would be reduced by pretreatment with histamine
H1- and
H2-receptor blocking agents and
with cyclooxygenase and lipoxygenase pathway inhibitors.
Anesthetized animals were examined on separate occasions 3 wk apart in
which one of the treatments was randomly given. The pulmonary arterial
occlusion technique was used to determine segmental pressure drops.
During ragweed challenge, PVR increased 
4 times compared with the
preshock value (3.04 vs. 12.07 mmHg · l
1 · min;P < 0.05). Although both Rus and Rds increased postshock, the greatest
relative increase occurred in Rds. None of the treatments reduced
partitioned resistances compared with no treatment. Our results show
that, under conditions of anaphylactic shock, increases in Rus and Rds
could not be ascribed to release of histamine or products of the
cyclooxygenase and lipoxygenase pathways.
histamine; asthma; thromboxane; leukotrienes
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
IN ANAPHYLACTIC SHOCK, the release from basophils and
mast cells of preformed and newly generated mediators leads to
pulmonary vasoconstriction (19, 26, 32). Among others,
histamine and products of the cyclooxygenase [i.e., prostacyclin,
prostaglandin (PG) F2
, and
thromboxane (Tx)] and lipoxygenase pathways [leukotrienes
(LTs): LTB4,
LTC4,
LTD4,
LTE4] may all modulate pulmonary vascular resistance (PVR) (1-4, 9, 17, 20, 24, 29,
30). Histamine, Tx,
PGF2, and the sulfidopeptide LTs cause vasoconstriction (1, 2, 4, 20, 24, 30), whereas prostacyclin
causes pulmonary vasodilation (17, 29). These mediators may also act on
different segments of the pulmonary vasculature. Histamine causes
predominantly postcapillary vasoconstriction (2, 8), whereas
LTB4 may cause precapillary
constriction (20). Effects of mediators on the pulmonary vasculature
may also be species specific (4).
In most studies, the effects of mediators on PVR have been examined under nonanaphylactic conditions (1, 2, 4, 20, 24, 30). In those instances, one or more mediators of anaphylaxis have been intravenously infused in in vitro or intact preparations at arbitrary concentrations after which changes in segmental pulmonary vascular resistances were determined. However, the relevance of such preparations to anaphylactic shock, in which an array of mediators is released during allergen challenge, is not clear.
In the present study, we used a ragweed model of anaphylactic shock (9, 18, 19) to determine the effect of allergen challenge on segmental pulmonary vascular resistances in which PVR was partitioned into an arterial upstream component (Rus) and a downstream (Rds) venous and capillary component (25). We further determined whether changes in Rus and Rds could be modified when pretreatment was undertaken with histamine H1-receptor blockade, histamine H2-receptor blockade, and cyclooxygenase and lipoxygenase pathway inhibition in this anaphylactic model.
| |
METHODS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Anaphylaxis model.
This protocol was approved by the University Animal Care Committee. Our
canine model of ragweed anaphylaxis has previously been described (18,
19). Newborn dogs received their first dose of antigen (0.5 mg ragweed
pollen extract) mixed with 30 mg
Al(OH)3 intraperitoneally within
24 h after birth. Injections were repeated at weekly intervals for 8 wk, at biweekly intervals for ~30 wk, and then at monthly intervals.
After 8 wk, some animals received airway challenges with ragweed (under
pentobarbital anesthesia, 30 mg/kg) rather than injections to maintain
hypersensitivity. This protocol was previously described in detail (3).
Both sensitizing regimens result in mean immunoglobulin E (IgE)
anti-ragweed antibody titers of >256 dilutions when measured 2 wk
before study by passive cutaneous anaphylaxis, whereas nonsensitized
littermate controls show no IgE titers by this test. There were no
differences in the anaphylactic response between the different
sensitization methods so that the animals were analyzed as a single
group. The animals were examined at ~1 yr of age.
Anaphylaxis protocol and measurements. Five treatment studies were performed on each animal ~3 wk apart in a randomized design. In the control treatment study (n = 9), no treatment was given. In the histamine H1-receptor-blocker study (n = 7), chlorpheniramine maleate (10 mg/kg iv) was administered (26); in the histamine H2-receptor-blocker study (n = 6), the animals were pretreated with ranitidine HCl (20 mg/kg iv) (18); in the cyclooxygenase-inhibition study (n = 7), this pathway was inhibited by indomethacin (2 mg/kg iv) (15); and in the lipoxygenase-inhibition study (n = 5), this pathway was inhibited by the 5-lipoxygenase-activating protein antagonist MK-0591 (see below) (5). Although in all dogs a control treatment study was performed, all treatments could not be performed in each dog; four dogs were withdrawn at various intervals for use in another study. Of the nine dogs studied, five completed the entire five treatments. However, the results were the same whether these five dogs were analyzed separately or whether the entire group of nine with missing values was analyzed.
During each treatment, four conditions were examined that included baseline, treatment, shock, and after intravenous volume expansion to return left ventricular end-diastolic pressure (LVEDP) to the preshock value (see further rationale in Data analysis). Previous investigators have shown that PVR is affected by the extent of pulmonary vascular hydration (33). Because LVEDP would fall during shock, volume was given to return LVEDP to the preshock value so that PVR and segmental resistances obtained pre- and postshock PVR could be compared at similar LVEDP. In the protocol, after baseline parameters were determined, one of the treatments was administered over 20 min, except for MK-0591. MK-0591 was administered initially as a 2 mg/kg iv bolus and then as an 8 µg · kg1 · min1
constant intravenous infusion for the remainder of the experiment and
was graciously supplied by Dr. A. W. Ford-Hutchinson of Merck Frosst
Canada, Inc. (5). In the control treatment
study in which no treatment was administered, placebo
(normal saline solution; 500 ml) was administered over this
interval. After an additional 40-min wait, ragweed antigen
(Ragweed Mix, catalogue no. 2315JW, Miles Pharmaceutical, Etiobicoke,
ON, Canada) was given to produce shock. On the basis of preliminary
experiments, we knew the approximate dose of allergen that produced
shock in each animal. Shock was defined as an ~50% fall in mean
aortic blood pressure (BP) from that found preshock. If, in a
particular study, this dose of antigen was not sufficient to produce
shock, then the antigen dose was doubled until shock occurred because
the presence of shock was the primary end point of the study. After
shock measurements were obtained, 10% pentastarch in normal saline
solution (500-600 ml) was given to return LVEDP back to baseline
condition.
Systemic and pulmonary hemodynamics were determined while the animals
were anesthetized with pentobarbital sodium (30 mg/kg iv; Ref.
28). The animal was ventilated (12 ml/kg; Harvard
Apparatus, Natick, MA) in the supine position with the rate adjusted as
necessary to maintain pH between 7.3 and 7.4. Supplemental oxygen was
given to maintain arterial PO2 >100
Torr throughout the study.
All vascular catheters were placed under sterile conditions. BP was
measured with a polyethylene catheter inserted into the femoral artery.
A Swan-Ganz catheter was advanced into the pulmonary artery to measure
mean pulmonary arterial pressure (Ppa), mean pulmonary wedge pressure
(Ppw), and right atrial pressure and to determine pulmonary capillary
occlusion pressure (Ppco; see Data
analysis). Cardiac output (CO) was
determined by thermodilution techniques (Columbus Instruments), with
the average of four determinations reported. All of the fluid-filled
catheters were connected to transducers (Cobe Instruments) and were
referenced relative to the left atrium. The left atrium
was determined as the lower one-third distance between the spine and
sternum on the basis of previous studies (22). All signals were
displayed on an eight-channel recorder (Astro-Med, West Warwick, RI).
Measurements were obtained at end expiration so that respiratory
variation would not affect interpretation of the results.
LVEDP was used to estimate venous downstream pressure (i.e., left
atrial pressure; Pla) because at end diastole we have observed little
difference between mean Ppw and LVEDP (11) and because others have
shown little difference between Pla and Ppw under normal conditions
(13) (see rationale in Data
analysis). A high-fidelity transducer-tipped catheter (Millar Instruments, Houston, TX) was advanced through a carotid artery incision and positioned into the left
ventricle (LV). Because no "a" wave is observed at the high heart
rates found in the present study, LVEDP was defined as the pressure at
which the rate of pressure development increased by 150 mmHg/s and was
sustained for at least 50 ms (11, 28, 31).
During each condition, plasma concentrations of histamine, LTs
(LTE4), as well as the breakdown
products of prostacyclin and TxA2
(i.e., 6-keto-PGF1 and
TxB2, respectively) were
obtained. It is recognized that many other mediators may
be released during shock, but these were considered the important ones
in terms of the present study. Mediators were measured by
radioimmunoassay techniques. Histamine immunoanalysis (Immunotech
International) was performed by competition between modified histamine
in the sample and the iodinated histamine tracer for the binding to the antibody coated on tubes (23).
6-Keto-PGF1,
TxB2, and
LTE4 were measured by NEN Research
Products (Boston, MA) NEK-008, NEK-007, and NEK-043, respectively (15).
For each mediator, the standard curves were generated in which a known
amount of the tracer was placed into an aliquot of pooled canine plasma
and not just the buffer solution alone. In the LT assay, because in
vitro conversion of LTC4 and
LTD4 to
LTE4 occurs spontaneously,
measurements of all LTs were obtained after enzymatic conversion (by
-glutamyl transpeptidase and microsomal leucine aminopeptidase) to
LTE4 as described by Heavey et al.
(16). For all mediators, samples were stored at
70°C and analyzed in duplicate.
Supplemental protocols. Other protocols were performed as a part of this study to control for the effects of time and other methodological concerns on our measurements. In the volume-infusion time-control protocol, volume infusion was not given after shock, and another set of shock measurements was obtained over this interval. The objective of this study was to determine what changes in partitioned resistances would occur over this interval without the effect of volume infusion.
In the sham-shock protocol, we controlled for the effects of time and the ragweed diluent (i.e., normal saline solution) on our hemodynamic measurements. The ragweed diluent alone was administered during the interval of antigen challenge, and there was no need for volume infusion. In the volume-depleted control study, the objective was to examine what the changes in PVR and partitioned resistances would be when, in nonshocked dogs, LVEDP and CO were lowered to an extent found in the anaphylaxis protocol. As previously stated, PVR is dependent on the extent of hydration (33), and, because LVEDP during allergen challenge would fall compared with the preshock value, this effect would by itself increase PVR and segmental resistances. The magnitude of this effect was determined in the volume-depleted control study. Pulmonary hemodynamics were determined when the animals were slightly dehydrated and maintained without water for 18 h before being studied. This gave LVEDP and CO values comparable with values found in the anaphylaxis protocol. In the lung mechanics study, the objective was to determine whether the increases in PVR and other resistances observed during allergen challenge may be related to bronchoconstriction and accompanying hyperinflation. In that case, extravascular factors, such as increases in end-expiratory pleural pressure (Ppl) and intrinsic positive end-expiratory pressure (PEEP) [i.e., alveolar pressure (PA) >0 cmH2O measured at end-expiration with the expiratory port of the ventilator occluded] may have contributed to the higher Ppa found during challenge. Lung mechanics were measured on a separate occasion at the end of the anaphylactic protocol in five sensitized dogs under pentobarbital anesthesia (30 mg/kg iv). The animals were placed into a volume-displacement plethysmograph as previously described (21). Lung volume was measured by a Krogh spirometer mounted on the plethysmograph. Flow was estimated from a pneumotachograph (model 3800, Hans Rudolph, Kansas City, MO) placed between the spirometer and the plethysmograph. Airway opening pressure (Pao) was obtained from a lateral pressure tap placed in the endotracheal tube. The lateral pressure tap was connected to a Validyne pressure transducer (Validyne, Northridge, CA). Ppl was estimated by an esophageal balloon, which was placed 5 cm from the gastroesophageal junction and connected by polyethylene tubing to the other port of the Validyne pressure transducer. The output of the pressure transducer could be displayed as Pao, Ppl, or transpulmonary pressure (Ptp). Signals for flow, volume, and pressure were displayed on the oscillograph. During tidal inflation, lung compliance (CL; ml/cmH2O) was measured from change in lung volume/
Ptp. The animals were studied at
baseline, during shock, and after volume infusion as described in the
anaphylaxis study.
Data analysis.
In the different protocols, PVR was calculated as (Ppa LVEDP)/CO. The rationale for the use of LVEDP rather than Ppw was as
follows. Before allergen challenge, there was no difference between
LVEDP and Ppw (see RESULTS), and
thus both could be used interchangeably to calculate PVR. However,
during allergen challenge, the time course of the Ppa decay was so slow
(>5 s; Fig. 1) that frequently Ppw > LVEDP. This indicated that a true LV filling pressure was never
reached, and thus use of LVEDP rather than Ppw represented a more
accurate picture of PVR (see
DISCUSSION).
|
Ppco)/CO; Rds, which predominantly represented venous and capillary
resistance, was calculated from (Ppco LVEDP)/CO (25).
Statistics. A one-way analysis of variance for repeated measures with missing numbers (ANOVA1R) was used when conditions for a specific treatment in the anaphylaxis protocol or conditions in the respective supplementary protocols were compared. ANOVA1R was also used when a given condition (i.e., shock, volume infusion) was compared between treatments and interventions in the anaphylaxis and supplementary protocols. When multiple means were compared, a Duncan's multiple-comparison test was used. Results are reported as means ± SD.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
In the anaphylaxis protocol, the hemodynamic measurements obtained in
the control treatment study are shown in Table
1. At baseline, LVEDP was
nearly identical to Ppw, and Ppco was just slightly greater than Ppw.
In Fig. 1A, the Ppa decline found
during occlusion was analyzed in terms of a biexponential model. After occlusion, there was a initial rapid decline in Ppa and a plateau in
pressure (i.e., Ppw) was reached in 2 s. PVR averaged ~2
mmHg · l1 · min
(Fig. 2). Because Ppco was calculated to be
very close to Ppw (Table 1), Rus accounted for most of PVR (Figs.
3 and 5), whereas the contribution of Rds
was fairly small (Figs. 4 and 6).
|
|
|
|
When antigen was given in the control treatment study,
mean BP fell 50% during shock (Table 1), and PVR increased
approximately fourfold (Fig. 2). During shock, Ppw (Table 1) was often
higher than LVEDP, and thus Ppw did not represent LV filling pressure in many experiments. As shown in Fig.
1B, the profile of occlusion pressure
during shock was very slow. Although both Rus and Rds increased during shock compared with preshock values (Figs. 3 and 4,
respectively), the relative increase in Rds was much greater than in
Rus (Figs. 5 and
6 in the control treatment
study).
|
|
In the control treatment study, slightly more volume was given than what was intended so that LVEDP and Ppa obtained post-volume infusion were slightly higher than preshock values (Table 1). In general, however, during volume infusion, indexes of pulmonary hemodynamics returned to preshock values, although the distribution of resistance still favored a relative increase in Rds (Fig. 6). During volume infusion, the return in PVR (Fig. 2) toward preshock values was not due to an effect of time alone because, in the volume-infusion protocol (Table 2), the changes in parameters over this interval were much less than what was observed in the control treatment study (compare volume condition found in Figs. 2-6 with the sham volume-infusion values in Table 2).
|
Moreover, in the sham-shock protocol, ragweed diluent was administered during the "shock interval," and this study also served as a time control study. In the sham-shock protocol, there was no decrease in BP when the diluent was administered. BP measured 152 ± 9 mmHg at baseline, 148 ± 10 mmHg at sham shock, and 157 ± 13 mmHg during volume infusion.
In the control treatment study (n = 9), the dose of antigen required to produce shock in the individual dogs (Table 1) ranged between 0.1 and 85 mg and averaged 25 ± 28 mg. Only in the histamine H1-receptor study (n = 7) was there a different antigen dose required to produce shock which doubled to 55 ± 50 mg (P < 0.05 vs. control study). In the histamine H2-receptor study (n = 6), the mean dose was 35 ± 16 (SD) mg; in the cyclooxygenase inhibition study (n = 7), the mean dose was 34 ± 20 mg; and in the lipoxygenase inhibition study (n = 5), the mean dose was 31 ± 15 mg.
The hemodynamic measurements obtained with the different treatments are shown in Tables 3-6. Relative to baseline values, hemodynamics were unchanged when any of the treatments was administered. Although, by design, BP fell approximately one-half during shock compared with preshock value, in the histamine H2-receptor study (Table 4), the fall in BP occurred to the greatest extent, and BP was significantly less than that found in either the histamine H1-receptor study (Table 3) or the cyclooxygenase inhibition study (Table 5).
|
|
|
|
In the different treatment studies, the increases in PVR (Fig. 2), Rus (Fig. 3), and Rds (Fig. 4) found between pre- and postshock conditions were similar to those in the control treatment study, although in the cyclooxygenase inhibition study, during volume infusion, Rds as well as percent Rds/PVR were significantly lower than values found in the control treatment study (Figs. 4 and 6).
The concentration of mediators found in the different studies are shown
in Table 7. In the histamine
H1-receptor study,
the concentration of histamine released during shock was higher than values in the other studies, and this finding reflected the higher antigen dose given in the histamine
H1-receptor study.
In the cyclooxygenase-inhibition study, the
concentrations of TxB2 and 6-keto-PGF1 were lower, whereas
LTE4 concentrations were higher,
than values in the other studies.
|
In the volume-depleted protocol, LVEDP averaged 2.2 ± 2.1 mmHg, Ppw averaged 2.6 ± 2 mmHg, and CO averaged 1.5 ± 1 l/min. These values compared well with those found in Table 1.
However, there was a different pattern of response in PVR, Rus, and Rds between what was found in the anaphylaxis protocol and
the volume-depleted protocol. In the volume-depleted
protocol, PVR during volume depletion (4.26 ± 1.81 mmHg · l1 · min)
increased slightly (but not significantly;
P < 0.11) compared with baseline
(PVR = 3.0 ± 0.94 mmHg · l1 · min),
but the increase was much smaller than that found between baseline and
shock in the control treatment study
(P < 0.05 between control and volume-depleted studies).
Moreover, in the volume-depleted protocol, although
both Rus (3.49 ± 1.39 mmHg · l1 · min;
P < 0.11) and Rds (0.78 ± 0.74 mmHg · l1 · min;
P < 0.01) were higher compared with
baseline values (2.42 ± 0.94 and 0.58 ± 0.83 mmHg · l1 · min,
respectively), most of the increase was in Rus rather than in Rds. With
volume depletion, Rus represented 84 ± 13% of PVR whereas Rds
represented 16 ± 13% of PVR, and these values were not different
from baseline percentages (78 ± 18 and 22 ± 19%,
respectively). In the volume-depleted control study,
percent Rus/PVR was significantly higher, whereas percent Rds/PVR was lower, than corresponding values found during shock in the control study.
Finally, in the lung mechanics study, the results showed that air trapping did not occur at end expiration. At end expiration, Ppl and Ptp were unchanged in the three conditions, and intrinsic PEEP was not present (Table 8). There were no changes in Ppl or PA during breath holding when measurements were obtained in any of the conditions. Although peak Ptp increased during shock, which indicated that bronchoconstriction was present, there was no change in CL (either static or dynamic) during allergen challenge.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
In our ragweed model, the results showed that during allergen challenge PVR increased approximately fourfold compared with the preshock measurement. Although both Rus and Rds increased during shock, the greatest relative increase occurred in Rds. With volume hydration, Rus and Rds returned toward preshock values, but percent Rds/PVR still remained higher than that found at baseline.
During shock, the increases in PVR and partitioned resistances observed in our model were not simply due to a passive effect of lower flows resulting from a fall in LVEDP (33). During allergen challenge, as LVEDP fell, zone II conditions of West et al. (33) (Ppa > PA > LVEDP) would evolve in the upper lung regions. PVR and segmental resistances would increase due to zonal changes as flow decreased during shock. The magnitude of this contribution was examined in the volume-depletion protocol in which the distribution of zones of West et al. and CO were similar to values found during anaphylaxis. We found that, in anaphylactic shock, the increase in PVR was much greater than that found during volume depletion and that the primary increase in resistance was in Rds. In contrast, in the volume-depletion model, the primary increase in resistance was in Rus.
The increase in pulmonary vascular pressures in the anaphylactic protocol was also not due to air trapping because there was no evidence to suggest that, at end expiration, increases in Ptp, Ppl, or intrinsic PEEP were present during allergen challenge (Table 8). In all conditions, expiratory flows at end expiration returned to zero, and there were no changes in Ppl and PA during breath holding when measurements were obtained. This would agree with our previous study in which histamine aerosolization failed to produce air trapping in a canine model of severe bronchoconstriction (12). Furthermore, it is not possible to exclude that perivascular interstitial edema caused by release of inflammatory mediators from mast cells led to an increase in vasoconstriction by means of compression of the pulmonary vasculature. However, this explanation appears unlikely as the sole cause of our results because pulmonary vasoconstriction occurred within seconds of allergen challenge and was reversed within minutes by intravascular volume expansion.
We considered that the most likely mechanism for the increase in PVR found during shock was related to a direct effect of mediators on the pulmonary circulation (1, 2, 4, 20, 22, 24, 30). Histamine has been shown to cause vasoconstriction, in canine lungs, that has been most pronounced in Rds (2, 4, 14). This has been shown in many preparations, including isogravimetric and double-occlusion determinations of pulmonary capillary pressure (7, 8, 14). Because histamine was released in large quantities in our model (Table 7), we expected that the increase in Rds would be attenuated by histamine H1- and H2-receptor blockade. However, we did not find that H1- or H2-receptor blockade modified the increases in PVR or partitioned resistances found in our model.
Such results support those of Silverman et al. (26), who examined the effects of antihistamines on cardiopulmonary changes in an Ascaris suum canine model of anaphylaxis. During challenge, H1-receptor blockade, H2-receptor blockade, or combined H1- and H2-receptor blockade, had no effect on PVR in their model. Silverman et al. hypothesized that one possibility for this finding was that the local concentrations of histamine were too high for the dose of chlorpheniramine used during shock in their model.
Although an increase in the dose of chlorpheniramine may produce a greater degree of histamine H1-receptor blockade during challenge, we found that an unwanted effect of this higher dose was that a higher allergen dose was needed to produce shock (18) because H1-receptor blockers inhibit mediator release from mast cells (27). With a higher antigen dose, more histamine was released during challenge in the H1-receptor blocker study than in the control study (Table 7). At constant antigen dose in our previous study (18), mediator release and shock under histamine H1-receptor blockade were attenuated compared with no treatment. Thus, in the present study and in the Ascaris model, a demonstrated effect of histamine on PVR during shock remains difficult to discern.
In the cyclooxygenase inhibition study, indomethacin was infused to suppress both the vasodilating (i.e., prostacyclin) and vasoconstricting (i.e., Tx) products indigenous to this pathway (2, 17). In turn, the net effect on these mediators on PVR would depend on the relative contribution of each. Our results showed that cyclooxygenase inhibition did not significantly alter partitioned resistances, although Rus and Rds were slightly lower than those found in the control treatment study.
With volume infusion, however, Rds and percent Rds/PVR measured in the cyclooxygenase inhibition study decreased slightly compared with the control treatment study values, although these differences were small. Barman and Taylor (2) showed that the Tx analog U-46619 produced a predominant increase in PVR in the large and small veins, although bronchoconstriction was not produced in their preparation. Our results suggest that Tx contributed to the increase in Rds observed during shock, but this effect was only evident post-volume infusion.
In the lipoxygenase inhibition study, the cysteinyl-containing derivatives LTC4, LTD4, and LTE4 are mainly vasoconstrictors and MK-0591 was administered to inhibit LT biosynthesis (5). In isolated perfused guinea pig lungs, Noonan et al. (24) found that infusion of either LTC4 or LTD4 caused an increase in PVR that occurred predominantly in the venous segment. In the present study, although MK-591 prevented the increase in LTs otherwise found during shock, we could not find that partitioned resistances decreased with lipoxygenase inhibition.
Although many mediators are released during anaphylaxis (i.e., among others, platelet-activating factor, eosinophil and neutrophil chemotactic factors, cytokines, etc.; Refs. 6, 32), in terms of the objectives of the present study, we concentrated on those mediators that, on the basis of previous experiments, have assumed important roles in causing pulmonary vasoconstriction in anaphylaxis (1, 17, 20, 24, 30). Most work on mechanisms of pulmonary hypertension in anaphylaxis has been based on simulated models of this condition (1, 20, 24). However, in this in vivo model of anaphylaxis, we could not relate pulmonary vasoconstriction to histamine release or products of the cyclooxygenase or lipoxygenase pathways.
Interestingly, however, our results showed that, during volume
infusion, PVR and partitioned resistances decreased in all treatments
compared with shock values and also compared with the volume-infusion
time-control study in which volume was not infused over
this interval. Once shock is produced in our model, there is a general
recovery of hemodynamics over time, although PVR does not usually
return to the preshock value for 2 h (9). Mediators are metabolized
postshock and would further be diluted by volume expansion. Moreover,
with volume infusion, pulmonary vascular pressure would increase, and
this could provide greater stretch on vascular smooth muscle. By
reversal of the bronchoconstrictive effect of the mediators, an
increase in pulmonary vascular pressures would return PVR toward
preshock values at a faster rate than without volume infusion.
Alternatively, saline infusion could affect PVR by altering blood
viscosity and by affecting cytokines, vasodilator peptides, or other
mediators not measured in the present study.
We also recognize multiple techniques could be used to determine Ppco in our model (7, 8, 10, 25). The gold standard is probably the double-occlusion technique (14) in which the pulmonary artery or vein can be acutely occluded to obtain the respective pressure drops across the arterial, venous, and middle segments. However, this technique would be difficult to use in our chronic in vivo preparation. Because multiple treatments were performed in the same animal over a 4- to 5-mo period, it would be difficult to implant a catheter into one of the pulmonary veins over this time period and to keep the animal healthy.
Furthermore, many mathematical models have been proposed to explain the decay in arterial occlusion pressure (8), which has been explained in terms of monoexponential or biexponential equations. In addition, many approaches have been offered as to when the occlusion should be initiated relative to the arterial pressure tracing. Gilbert and Hakim (10) found little difference in the results when occlusion was initiated between 0 and 150 ms after an initial pressure change was noted and further showed that the extrapolated exponent was relatively insensitive to the exact number of data points used, for instance from 0.2 to 1.5 s or from 0.2 to 2.5 s. In the present study, we chose the time at which a pressure change was first detected and tried to include data obtained over 5-8 s because the pressure decay was so slow once shock was produced.
In the different treatment studies, we found that at baseline Rus
accounted for ~70% of PVR whereas Rds accounted for 30%. In the
literature, a wide range of values has been reported for these
fractions with Rds ranging between 30 and 50% of PVR. Much of
variability may depend on the preparation used, open- or closed-chest, isolated lobe, zones of West et al. (33), etc. (7, 8, 14). In a
closed-chest canine preparation with similar methodology, Cope et al.
(7) reported that at baseline Rus averaged 71% of PVR whereas Rds
averaged 29% of PVR. They also did not find much of a difference in
parameters when determined by graphic or visual fit.
In the control treatment study, we compared values of
Ppco obtained by best-visual-fit line with those in which the slow
component of pressure decay was fitted to an exponential function by
regression analysis. Between the two methods, the coefficient of
correlation was >0.98 in all conditions (0.99 at baseline, 0.995 at
shock, and 0.985 during volume infusion) with a slope of nearly 1
(1.06 at baseline, 1.08 during shock, and 1.08 during volume infusion) and with a y-intercept of nearly zero
(0.7 vs. 3.5 vs. 0.38 cmH2O in which the visual method
is plotted on the ordinate). Because the methods gave similar results,
we do not think that different ways of analyzing the data would have
yielded conclusions different from those already discussed.
In the anaphylactic protocol, another point to consider
in our measurement of Ppco is that, during the five treatment studies, occlusion of vessels in different parts of the lung may have affected calculation of segmental vascular resistances in the various
conditions. In a previous study (22), it was noted that
the pulmonary catheter always occluded one of the lower lobes during
inflation in the dog. We could not check the catheter location by
fluoroscopy under the different conditions to ensure that it was at a
constant location. However, preshock measurements were obtained in the
same animal multiple times at baseline (5) before any treatment was
given. Accordingly, we could compare the percent coefficient of
variation [CV = (SD/mean value) × 100%] of the
segmental resistances measured (i.e., percent Rus) to determine the
extent to which variability may have contributed to our results. For
all dogs, percent CV was 14% with a range of 9.7-29%. Moreover,
when the percent Rus measurement was repeated in the control treatment
study ~1 h apart, the percent CV was also small at
15%. In contrast, because the changes in percent Rus observed during
shock were so large, we do not believe that measurement variability in
partitioned resistances, possibly reflecting different positions of the
catheter, altered the interpretation of our results.
We further recognize that to some extent PVR may be dependent on flow and that differences in CO may have confounded interpretation of changes in PVR, Rus, and Rds that were found pre- vs. post-volume infusion (33). In the control treatment study, we found that, post-volume infusion, the mean CO was slightly higher than the preshock value (Table 1). However, in other treatment studies, for instance, in the lipoxygenase treatment study, CO values were the same at baseline and post-volume infusion (Table 6), and similar findings in PVR, Rus, and Rds were obtained.
Our study also indicates that Ppw may not reflect LV filling pressure during anaphylaxis. Because the decay in arterial occlusion pressure may be so slow, a true Pla may not be observed. In the present study, we used LVEDP to reflect mean atrial pressure. Previous results showed that, under normal conditions, there was good agreement between these Ppw and LVEDP (11), and other authors have reported no difference between Ppw and Pla under normal conditions (13). It would be difficult to stop the ventilator for the length necessary to ensure that a true Ppw had been reached in the present study. Our results show, therefore, that, in the clinical situation of anaphylactic shock, one must be careful in the interpretation of Ppw as representing LV filling pressure.
In summary, our results show that, in a ragweed model, there was a large increase in PVR observed during shock and that the largest relative increase in resistance occurred in the venous and capillary segments. None of the treatments used in this model could reverse pulmonary vasoconstriction during shock: either the local concentration of mediators released during challenge were too great for the doses used, combinations of treatments were required, or mediators not thus far examined were more important in causing vasoconstriction. Although previous studies in nonallergic preparations (1, 2, 4, 20, 24, 30) have indicated that histamine release and products of the lipoxygenase and cyclooxygenase pathways may be important in producing pulmonary vasoconstriction in anaphylaxis, none of these treatments were able to attenuate the pulmonary vascular effects found in this in vivo allergic model.
| |
ACKNOWLEDGEMENTS |
|---|
This study was supported by the Manitoba Heart and Stroke Foundation.
| |
FOOTNOTES |
|---|
Address for reprint requests: S. N. Mink, GF-221, Health Science Center, 700 William Ave., Winnipeg, Manitoba, Canada R3E OZ3.
Received 20 December 1996; accepted in final form 30 October 1997.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
1.
Ahmed, T.,
B. Marchette,
A. Wannner,
and
L. Yerger.
Direct and indirect effects of leukotriene D4 on the pulmonary systemic circulations.
Am. Rev. Respir. Dis.
131:
554-558,
1985[Medline].
2.
Barman, S. A.,
and
A. E. Taylor.
Histamine's effect on pulmonary vascular resistance and compliance at elevated tone.
Am. J. Physiol.
257 (Heart Circ. Physiol. 26):
H618-H625,
1989
3.
Becker, A. B.,
J. Hershkovich,
F. E. R. Simons,
K. J. Simons,
M. K. Lilley,
and
M. W. Kepron.
Development of chronic airway hyperresponsiveness in ragweed-sensitized dogs.
J. Appl. Physiol.
66:
2691-2697,
1989
4.
Bradley, J. D.,
P. B. Zanaboni,
and
T. E. Dahms.
Species differences in pulmonary vasoactive responses to histamine, 5-hydroxytryptamine, and KCl.
J. Appl. Physiol.
74:
139-146,
1993
5.
Brideau, C.,
C. Chan,
S. Charleson,
D. Denis,
J. Evans,
W. Ford-Hutchinson,
R. Fortin,
J. Gillard,
D. Guevremont,
J. Hutchinson,
T. Jones,
S. Leger,
J. Mancini,
C. McFarlane,
C. Pickett,
H. Piechuta,
P. Prasit,
D. Riendeau,
C. Rouzer,
P. Tagari,
P. Vickers,
and
R. Young.
Pharmacology of MK-0591, a potent, orally active leukotriene biosynthesis inhibitor.
Can. J. Physiol. Pharmacol.
70:
799-807,
1992[Medline].
6.
Campbell, W.,
and
P. Halushka.
Lipid derived autoacoids: eicosanoids and platelet activating factor.
In: Goodman and Gilman's The Pharmacological Basis of Therapeutics (9th ed.), edited by J. Hardman,
and L. Limbird. New York: McGraw-Hill, 1996, p. 601-616.
7.
Cope, D. K.,
R. C. Allison,
J. L. Parmentier,
J. N. Miller,
and
A. E. Taylor.
Measurement of effective pulmonary capillary pressure using the pressure profile after pulmonary artery occlusion.
Crit. Care Med.
14:
16-22,
1986[Medline].
8.
Dawson, C. A.,
T. A. Bronikowski,
J. H. Linehan,
S. T. Haworth,
and
D. A. Rickaby.
On the estimation of pulmonary capillary pressure from arterial occlusion.
Am. Rev. Respir. Dis.
140:
1228-1236,
1989[Medline].
9.
Elkin, J.,
K. Duke,
G. Price,
W. Kepron,
A. Becker,
and
S. Mink.
Does bolus epinephrine improve systemic hemodynamics during anaphylactic shock in dogs (Abstract)?
Am. J. Respir. Crit. Care Med.
149:
A1060,
1994.
10.
Gilbert, E.,
and
T. S. Hakim.
Derivation of pulmonary capillary pressure from arterial occlusion in intact conditions.
Crit. Care Med.
22:
986-993,
1994[Medline].
11.
Gomez, A.,
H. Unruh,
and
S. N. Mink.
Altered left ventricular chamber stiffness and isovolumic relaxation in dogs with chronic pulmonary hypertension caused by emphysema.
Circulation
87:
247-260,
1993
12.
Greville, H. W.,
M. E. Arnup,
and
S. N. Mink.
Density dependence of maximal flow is lung volume dependent during bronchoconstriction.
J. Appl. Physiol.
62:
691-705,
1987
13.
Grossman, W.
Cardiac catherization.
In: Heart Disease, edited by E. Braunwald. Philadelphia, PA: Saunders, 1992, p. 180-203.
14.
Hakim, T. S.
Criteria for analysis of arterial and venous occlusion.
J. Appl. Physiol.
70:
665-675,
1991
15.
Hanly, P.,
A. Sienko,
and
R. B. Light.
Effect of cyclooxygenase blockade on gas exchange and hemodynamics in Pseudomonas pneumonia.
J. Appl. Physiol.
63:
1829-1836,
1987
16.
Heavey, D.,
R. Soberman,
R. Lewis,
B. Spur,
and
K. Austen.
Critical considerations in the development of an assay for sulfidopeptide leukotrienes in plasma.
Prostaglandins
33:
693-798,
1987[Medline].
17.
Hyman, A.,
and
P. J. Kadowitz.
Pulmonary vasodilator activity of prostacyclin (PGI2) in the cat.
Circ. Res.
45:
404-409,
1979
18.
Kepron, W., E. Correa, H. Unruh, K. Duke, and S. Mink.
Different effects of blockers of mediators of anaphylaxis on
cardiac hemodynamics during ragweed challenge in dogs (Abstract). Am. Rev. Respir. Dis. A779,
1991.
19.
Kepron, W.,
J. James,
B. Kirk,
A. Sehon,
and
K. Tse.
A canine model for reaginic hypersensitivity and allergic bronchoconstriction.
J. Allergy Clin. Immunol.
59:
64-69,
1977[Medline].
20.
Malik, A. B.,
T. C. Noonan,
W. M. Selig,
and
J. G. Garcia.
Effects of exogenous leukotrienes on the pulmonary circulation.
Prog. Clin. Biol. Res.
199:
221-235,
1985[Medline].
21.
Mink, S. N.,
J. J. Coalson,
L. Whitely,
H. Greville,
and
C. Jadue.
Pulmonary function tests in the detection of small airway obstruction in a canine model of bronchiolitis obliterans.
Am. Rev. Respir. Dis.
130:
1125-1133,
1984[Medline].
22.
Mink, S. N.,
R. B. Light,
T. Cooligan,
and
L. D. H. Wood.
Effect of PEEP on gas exchange and pulmonary perfusion in canine lobar pneumonia.
J. Appl. Physiol.
50:
517-523,
1981
23.
Morel, A. M.,
and
M. Delaage.
Immunoanalysis of histamine through a novel chemical derivatization.
J. Allergy Clin. Immunol.
82:
646-654,
1988[Medline].
24.
Noonan, T. C.,
W. M. Selig,
D. F. Kern,
and
A. B. Malik.
Mechanism of peptidoleukotriene-induced increases in pulmonary transvascular fluid filtration.
J. Appl. Physiol.
61:
1928-1934,
1986
25.
O'Leary, C. E.,
R. Fiori,
and
T. S. Hakim.
Perioperative distribution of pulmonary vascular resistance in patients undergoing coronary artery surgery.
Anesth. Analg.
82:
958-963,
1996[Abstract].
26.
Silverman, H.,
W. Taylor,
P. Smith,
A. Kagey-Sobotka,
S. Permutt,
L. Lichtenstein,
and
E. Bleecker.
Effect of antihistamines on the cardiopulmonary changes due to canine anaphylaxis.
J. Appl. Physiol.
64:
210-217,
1988
27.
Ting, S.,
D. Rauls,
and
B. Reiman.
Inhibitory effect of hydroxyzine on antigen-induced histamine release in vivo.
J. Allergy Clin. Immunol.
75:
63-65,
1985[Medline].
28.
Unruh, H.,
R. Wang,
D. Bose,
and
S. Mink.
Does pentobarbital anesthesia depress left ventricular contractility in dogs?
Am. J. Physiol.
261 (Heart Circ. Physiol. 30):
H700-H706,
1991
29.
Voelkel, N. F.,
J. G. Gerber,
I. F. McMurtry,
A. S. Nies,
and
J. T. Reeves.
Release of vasodilating prostaglandin, PGI2, from isolated rat lung during vasoconstriction.
Circ. Res.
48:
207-213,
1981
30.
Voelkel, N. F.,
R. Stenmark,
J. T. Reeves,
M. M. Mathias,
and
R. C. Murphy.
Actions of lipoxygenase metabolites in isolated rat lungs.
J. Appl. Physiol.
57:
860-867,
1984
31.
Walsh, R.,
and
R. O'Rourke.
Direct and indirect effects of calcium blocking agents on isovolumic left ventricular relaxation in conscious dogs.
J. Clin. Invest.
75:
1426-1434,
1985.
32.
Wasserman, S.,
and
D. Marquardt.
Anaphylaxis.
In: Allergy Principles and Practice (3rd ed.), edited by E. Middleton,
C. Reed,
E. Ellis,
N. Adkinson,
and J. Yunginer. St. Louis, MO: Mosby, 1988, p. 1365-1376.
33.
West, J. B.,
C. T. Dollery,
and
A. Naimark.
Distribution of blood flow in isolated lung; relation to vascular and alveolar pressure.
J. Appl. Physiol.
19:
713-724,
1964
This article has been cited by other articles:
|
|
 |
|
  C. CHRUSCH, S. SHARMA, H. UNRUH, E. BAUTISTA, K. DUKE, A. BECKER, W. KEPRON, and S. N. MINK Histamine H3 Receptor Blockade Improves Cardiac Function in Canine Anaphylaxis Am. J. Respir. Crit. Care Med., October 1, 1999; 160(4): 1142 - 1149. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Mink, A. Becker, S. Sharma, H. Unruh, K. Duke, and W. Kepron Role of autacoids in cardiovascular collapse in anaphylactic shock in anesthetized dogs Cardiovasc Res, July 1, 1999; 43(1): 173 - 182. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P < 0.05 vs. preshock;
!! P < 0.05 cyclooxygenase study vs. control
study.