Professional Documents
Culture Documents
Hypercapnic Acidosis and Mortality in Acute Lung Injury
Hypercapnic Acidosis and Mortality in Acute Lung Injury
Figure 1. Day 1 pH and PaCO2 values of the study population. HA, hypercapnic acidosis; RA, respiratory
alkalosis; PBW, predicted body weight.
arterial blood gas data collected on the morning following randomization and institution of
mechanical ventilation according to the study
protocol (day 1). Although they reflect a single
point in time, day 1 data have the advantage of
being the most complete set of blood gas values and avoid the survival bias of requiring
serial measurements to define HA. The data
collection protocol provided day 1 blood gas
values from 7 hrs to as much as 31 hrs after
institution of protocolized mechanical ventilation. The day 1 blood gas values and definitions of HA and respiratory alkalosis are
shown in Figure 1. Because there is no consensus definition of therapeutic HA, our analysis plan included assessing the effect of different degrees of HA. These were defined a
priori. HA1 was defined as a pH 7.40 and
PaCO2 40 mm to assess the effect of any
degree of HA. HA2 was defined as a pH 7.35
and PaCO2 45 mm Hg to exclude patients
whose blood gas values would otherwise have
been considered within normal limits and to
assess a more severe degree of HA. There were
too few patients with greater extremes of respiratory acidosis (e.g., pH 7.30 and PaCO2
50 mm Hg) to perform reliable analyses.
Finally, we hypothesized a priori that extremes of acidosis, even if associated with elevated PaCO2, would indicate significant metabolic acidosis, comorbidities, and/or probable
inevitable irreversibility of the underlying disease. These patients would likely have such
overwhelming cellular dysfunction that HA
would not afford protection (10) or would be
poorly tolerated. Furthermore, patients with a
pH 7.15 may have been given sodium bicarbonate as part of the trial at the discretion of
the providers, but this information was not
recorded. Thus, a third definition of HA (HA3)
with a lower limit of pH and upper limit of
PaCO2 consistent with a simple but moderately
severe acute respiratory acidosis was examined
(7.15 pH 7.35 and 45 mm Hg PaCO2
65 mm Hg). Respiratory alkalosis was defined
as pH 7.45 and PaCO2 35 mm Hg. Again,
there were too few patients with more extreme
respiratory alkalosis (e.g., pH 7.50 and PaCO2
30 mm Hg) to perform reliable multivariate
analysis.
Since no direct measure of a metabolic
acidosis was recorded for the majority of patients on day 1, arterial HCO3 values were
calculated from the arterial pH and PaCO2 val-
12 mL/kg
6 mL/kg
No.
Age, years
Female
APACHE III score
Risk factor for ALI, n (%)
Trauma
Sepsis
Multiple transfusions
Aspiration
Pneumonia
Other
Day 1 physiologic values
Minute ventilation, L/min
Respiratory rate, min1
Tidal volume, mL/kg
Pplat, cm H2O
pH
PaCO2, mm Hg
Base excess, mEq/L
FIO2
PEEP, cm H2O
PaO2, mm Hg
PaO2/FIO2
Mortality rate, %
429
52 18
0.41
84 28
432
51 17
0.40
81 28
37 (8)
111 (26)
15 (3)
62 (14)
156 (36)
48 (11)
54 (12)
117 (27)
10 (2)
65 (15)
144 (33)
42 (10)
12.6 4.6
16 6
11.8 0.8
33 9
7.41 0.07
35 8
2.1 5.7
0.51 0.18
8.6 3.7
77 19
176 76
40
12.9 3.6
28 7
6.2 0.9
25 7
7.38 0.08
40 10
1.6 5.4
0.57 0.19
9.6 3.7
76 23
158 73
31
APACHE, Acute Physiology and Chronic Health Evaluation; ALI, acute lung injury; Pplat, plateau
pressure; PEEP, positive end-expiratory pressure.
RESULTS
A total of 861 patients were studied in
the parent clinical trial; 141 were excluded due to missing arterial blood gas
on day 1, and 16 were excluded due to
missing data in other variables. The demographics of the study population are
shown in Table 1 and of the excluded
patients in Table 2. Patients excluded due
to missing data had lower mortality rate
and statistically significantly lower Pplat,
FIO2, and, in the 6 mL/kg tidal volume
group, lower PEEP compared with the
patients included in the analysis. The remaining patients in each tidal volume
group were characterized as hypercapnic
or nonhypercapnic based on the definitions of HA as described. The characteristics of these four groups for HA1 are
shown in Table 3. Similar data for HA2
and HA3 are shown in Tables 4 and 5,
respectively. Patients from each of the
ten ARDS Network centers were represented in the 12 mL/kg HA1 group.
The results of the logistic regression
models for each tidal volume group and
each definition of HA are shown in Figure
2. The odds ratio (OR) of mortality and
95% confidence intervals (CI) of the 12
mL/kg and 6 mL/kg tidal volume groups
based on the presence of HA1 were 0.32
(CI 0.13 0.79, p .013) and 1.07 (CI
0.611.90, p .808), respectively. InCrit Care Med 2006 Vol. 34, No. 1
cluding more severe and restrictive definitions of HA in the model led to lower
adjusted ORs in the 12 mL/kg tidal volume group (Fig. 2). For HA2, the adjusted
OR of mortality in the 12 mL/kg tidal
volume group was 0.14 (CI 0.03 0.70, p
.016). Finally, HA3 was associated with
the greatest reduction in adjusted OR of
mortality in the 12 mL/kg tidal volume
group (OR 0.06, CI 0.01 0.47, p
.008). None of the adjusted ORs for mortality related to HA in the 6 mL/kg tidal
volume group were 1 or reached statistical significance (Fig. 2). Lastly, respiratory alkalosis was not statistically significantly associated with mortality in either
tidal volume group.
The additions of gender, day 1 tidal
volume, and/or PEEP did not alter the
association between HA and mortality.
For example, the OR of mortality associated with PEEP in the 12 mL/kg tidal
volume group with HA2 was 1.09 (95% CI
0.951.25, p .229). HCO3 values calculated from blood gas data were highly
correlated with venous values for those in
whom data were available. (According to
the method of Bland and Altman, the
calculated arterial HCO3 concentrations
were 0.87 mmol/L lower on average than
the measured venous values with an SD of
2.14 mmol/L. The venous-arterial difference and SD are consistent with changes
DISCUSSION
This study demonstrates that HA is
associated with reduced mortality in patients receiving 12 mL/kg tidal volumes
and is not associated with an effect on
mortality in patients receiving 6 mL/kg
tidal volumes in the ARDS Network clinical trial. These results support the hypothesis that HA is protective, at least in
some patients, rather than simply a tolerated side effect in the management of
patients with ALI. These findings are consistent with the theory that ventilatorassociated lung injury is occurring to a
greater extent in the 12 mL/kg tidal volume group and that HA mitigates this
injury. They are also consistent with the
theory that the lung-protective ventilation strategy considerably reduces this
injury to the point that a protective effect
of HA is not detectable (20, 21).
The acutely injured lung in the midst
of an inflammatory response has proven
refractory to targeted molecular interventions designed to reduce inflammation. The only intervention that has
proven effective in reducing mortality is
the reduction in tidal volume to 6
mL/kg vs. 12 mL/kg demonstrated in the
ARDS Network Trial (6). Our findings are
consistent with the growing basic science
evidence supporting the protective effects
of HA in states of lung injury. The original concept of salutary acidosis, the pH
paradox, arose from models of ischemia
and reperfusion. It has been observed re3
Table 2. Characteristics of patients lacking day 1 blood gas data and therefore excluded from the
analysis
12 mL/kg
6 mL/kg
Variable
Included
Excluded
Included
Excluded
No.
Age, years
Female, %
APACHE III score
Risk factor for ALI/ARDS, n (%)
Trauma
Sepsis
Multiple transfusions
Aspiration
Pneumonia
Other
Day 1 physiologic values
Minute ventilation, L/min
Respiratory rate, min1
Tidal volume, mL/kg
Pplat, cm H2O
pH
PaCO2, mm Hg
Base excess, mEq/L
FIO2
PEEP, cm H2O
PaO2, mm Hg
PaO2/FIO2
Mortality rate, %
369
52 18
42
85 28
60
54 17
35
79 31
351
51 17
40
82 28
81
52 17
38
78 17
31 (8)
95 (26)
15 (4)
52 (14)
134 (36)
42 (11)
6 (10)
16 (27)
0 (0)
10 (17)
22 (37)
6 (10)
41 (12)
96 (27)
7 (2)
50 (14)
124 (35)
33 (9)
13 (16)
21 (26)
3 (4)
15 (19)
20 (25)
9 (11)
12.6 4.6
16 6
11.8 0.8
33 9
7.41 0.07
35 8
2.1 5.6
0.51 0.18
94
76 19
175 76
40
12.2 3.5
17 7
11.9 0.8
29 6a
12.9 3.6
29 7
6.2 0.9
25 7
7.38 0.08
40 10
1.6 5.4
0.57 0.19
10 4
76 23
157 73
32
0.47 0.13
83
33
13.1 3.6
26 9
6.3 0.6
22 5a
0.49 0.14a
8 3a
26
APACHE, Acute Physiology and Chronic Health Evaluation; ALI, acute lung injury; ARDS, acute
respiratory distress syndrome; Pplat, plateau pressure; PEEP, positive end-expiratory pressure.
a
p .05 compared with patients included in the model.
6 mL/kg
Variable
No HA1
HA1
No HA1
HA1
No.
Age, years
Female, %
APACHE III score
Risk factor for ALI/ARDS, n (%)
Trauma
Sepsis
Multiple transfusions
Aspiration
Pneumonia
Other
Day 1 physiologic values
Minute ventilation, L/min
Respiratory rate, min1
Tidal volume, mL/kg
Pplat, cm H2O
pH
PaCO2, mm Hg
Base excess, mEq/L
FIO2
PEEP, cm H2O
PaO2, mm Hg
PaO2/FIO2
Mortality rate, %
332
52 18
41
85 27
37
50 20
54
84 35
247
51 16
41
82 28
104
49 18
38
81 27
27 (8)
83 (25)
11 (3)
51 (15)
122 (37)
38 (11)
4 (11)
12 (32)
4 (11)
1 (3)
12 (32)
4 (11)
24 (10)
71 (29)
5 (2)
42 (17)
80 (32)
25 (10)
17 (16)
25 (24)
2 (2)
8 (8)
44 (42)
8 (8)
12.8 4.7
16 6
11.9 0.7
32 8
7.42 0.06
34 7
2.3 5.7
0.50 0.17
83
77 19
181 76
41
10.9 3.6a
16 7
11.4 1.5a
40 12a
7.34 0.08a
48 5a
0.2 4.4a
0.66 0.20a
12 5a
70 15a
126 53a
35
13.2 3.8
29 7
6.3 0.9
24 6
7.41 0.07
35 6
2.3 5.4
0.53 0.17
93
77 24
168 75
30
12.3 3.1a
30 7
6.0 0.9a
28 7a
7.31 0.08a
51 9a
0.0 4.8a
0.65 0.21a
11 4a
73 17
132 63a
35
APACHE, Acute Physiology and Chronic Health Evaluation; ALI, acute lung injury; ARDS, acute
respiratory distress syndrome; Pplat, plateau pressure; PEEP, positive end-expiratory pressure; HA1 is
defined as pH 7.40 and PaCO2 40 mm Hg.
a
p .05 compared with patients without HA1.
Table 4. Characteristics of model population by tidal volume and hypercapnic acidosis (HA)2
12 mL/kg
6 mL/kg
Variable
No HA2
HA2
No HA2
HA2
No.
Age, yrs
Female, %
APACHE III score
Risk factor for ALI/ARDS
Trauma
Sepsis
Multiple transfusions
Aspiration
Pneumonia
Other
Day 1 physiologic values
Minute ventilation, L/min
Respiratory rate, min1
Tidal volume, mL/kg
Pplat, cm H2O
pH
PaCO2, mm Hg
Base excess, mEq/L
FIO2
PEEP, cm H2O
PaO2, mm Hg
PaO2/FIO2
Mortality rate, %
356
52 18
41
85 27
13
44 17
69
91 47
298
52 17
40
82 28
53
46 16a
45
81 27
29 (8)
89 (25)
14 (4)
52 (15)
130 (37)
42 (12)
2 (15)
6 (46)
1 (8)
0 (0)
4 (31)
0 (0)
35 (12)
85 (29)
6 (2)
44 (15)
98 (33)
30 (10)
6 (11)
11 (21)
1 (2)
6 (11)
26 (49)
3 (6)
12.6 4.6
16 6
11.8 0.8
32 8
7.42 0.06
34 7
2.1 5.7
0.50 0.17
83
77 19
178 75
40
12.1 4.5
20 10a
10.8 2.0a
47 15a
7.27 0.10a
52 5a
1.5 6.0
0.75 0.21a
14 6a
63 13a
100 38a
31
13.0 3.7
29 7
6.3 0.9
24 6
7.40 0.07
37 7
2.0 5.3
0.55 0.18
94
77 23
164 75
31
12.2 2.7
32 6a
5.9 0.9a
29 6a
7.28 0.09a
56 10a
0.4 5.5a
0.69 0.20a
12 4a
71 15
119 51a
38
APACHE, Acute Physiology and Chronic Health Evaluation; ALI, acute lung injury; ARDS, acute
respiratory distress syndrome; Pplat, plateau pressure; PEEP, positive end-expiratory pressure; HA2 is
defined as pH 7.35 and PaCO2 45 mm Hg.
a
p .05 compared with patients without HA2.
Table 5. Characteristics of model population by tidal volume and hypercapnic acidosis (HA)3
12 mL/kg
6 mL/kg
Variable
No HA3
HA3
No HA3
HA3
No.
Age, years
Female, %
APACHE
Risk factor for ALI/ARDS, n (%)
Trauma
Sepsis
Multiple transfusions
Aspiration
Pneumonia
Other
Day 1 physiologic values
Minute ventilation, L/min
Respiratory rate, min1
Tidal volume, mL/kg
Pplat, cm H2O
pH
PaCO2, mm Hg
Base excess, mEq/L
FIO2
PEEP, cm H2O
PaO2, mm Hg
PaO2/FIO2
Mortality rate, %
358
52 18
41
85 27
11
44 18
82a
82 52
306
51 17
39
82 29
45
45 16a
44
78 24
29 (8)
91 (25)
14 (4)
52 (15)
130 (36)
42 (12)
2 (18)
4 (36)
1 (9)
0 (0)
4 (36)
0 (0)
35 (11)
86 (28)
6 (2)
45 (15)
104 (34)
30 (10)
6 (13)
10 (22)
1 (2)
5 (11)
20 (44)
3 (7)
12.7 4.6
16 6
11.8 0.8
32 8
7.42 0.07
35 8
2.2 5.7
0.51 0.17
83
77 19
178 75
41
10.7 3.1
17 8
10.7 3.1
44 14a
7.31 0.04a
51 6a
0.4 4.1
0.71 0.20a
13 6a
62 13a
103 40a
18
13.0 3.7
29 7
6.2 0.9
25 7
7.39 0.08
38 9
1.9 5.6
0.55 0.19
94
77 23
162 75
31
12.1 2.8
31 7a
5.9 0.8a
28 6a
7.30 0.05a
53 5a
0.2 3.6a
0.66 0.19a
11 3a
72 16
125 52a
33
APACHE, Acute Physiology and Chronic Health Evaluation; ALI, acute lung injury; ARDS, acute
respiratory distress syndrome; Pplat, plateau pressure; PEEP, positive end-expiratory pressure; HA3 is
defined as 7.15 pH 7.35 and 45 mm Hg PaCO2 65 mm Hg.
a
p .05 compared with patients without HA3.
while also excluding patients who transiently failed to meet these criteria. Nevertheless, the finding that HA on day 1 in
the 12 mL/kg arm of the study was associated with better survival and the evidence of increasing effect with increasing
degrees of HA suggests that some protection by HA is afforded early in the course
of ALI even if it is not sustained. Further
support for a protective effect of even a
short duration of HA is evident in many
animal studies, where a benefit from acidosis accrues in only several hours (13
22, 24, 26 32). Unmeasured confounding
variables are a universal problem with
multivariate analyses of cohort studies.
We did not have data on the use of sodium bicarbonate, accurate estimates of
CO2 production, or deadspace ventilation.
Since there were limitations in the tolerable limits of low pH in the ARDS Network trial, our ability to test the hypothesis that HA is beneficial is limited since
extremes were avoided. If greater extremes of HA had been allowed in this
study, it is possible that that either more
benefit or more harm may have been
detected. Most patients do not have a
pure respiratory acidosis, so some patients with HA had a combined respiratory and metabolic acidosis. In addition,
although we treated it as a single entity,
patients may develop HA by different
mechanisms including reduced minute
ventilation, increased CO2 production,
increased deadspace ventilation, and increased shunt fraction. Each of these
mechanisms may have varying degrees of
pathophysiological significance. These
variables need to be considered and quantified in future studies where HA exists in
low tidal volume groups. Nevertheless,
although inability to control for these
6
CONCLUSION
This secondary analysis of data from
the ARDS Network trial of lower tidal
volume provides evidence in support of
the theory that HA exerts a protective
effect in ALI. This effect appears to be
modulated by the ventilatory approach.
Because of the study limitations, it is
important to appreciate that, although
these clinical observations support a body
of basic science on the beneficial effects
of HA, they do not confirm them. Confirmation awaits further clinical studies and
an appropriately designed randomized
trial of HA coupled with lung-protective
ventilation in patients with ALI. Based on
the data available in this and other studies, we propose that greater extremes of
HA than those found in the ARDS Network trial of 12 vs. 6 mL/kg tidal volumes
would likely be needed to measure any
effect of HA on outcome in ALI patients
receiving lung protective ventilation. The
optimal dose, duration, and type (lowered
REFERENCES
1. Hickling KG, Walsh J, Henderson S, et al:
Low mortality rate in adult respiratory distress syndrome using low- volume, pressurelimited ventilation with permissive hypercapnia: A prospective study. Crit Care Med
1994; 22:1568 1578
2. Amato MB, Barbas CS, Medeiros DM, et al:
Beneficial effects of the open lung approach with low distending pressures in
ARDS: A prospective randomized study on
mechanical ventilation. Am J Respir Crit
Care Med 1995; 152:18351846
3. Gentilello LM, Anardi D, Mock C, et al: Permissive hypercapnia in trauma patients.
J Trauma 1995; 39:846 52
4. Amato MB, Barbas CS, Medeiros DM, et al:
Effect of a protective-ventilation strategy on
mortality in the acute respiratory distress
syndrome. N Engl J Med 1998; 338:347354
5. Ranieri VM, Suter PM, Tortorella C, et al:
Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: A randomized
controlled trial. JAMA 1999; 282:54 61
6. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute
lung injury and the acute respiratory distress
syndrome. The Acute Respiratory Distress
Syndrome Network. N Engl J Med 2000; 342:
13011308
7. Kitakaze M, Weisfeldt ML, Marban E: Acidosis during early reperfusion prevents myocardial stunning in perfused ferret hearts.
J Clin Invest 1988; 82:920 927
8. Orchard C, Kentish J: Effects of changes of
pH on the contractile function of cardiac
muscle. Am J Physiol 1990; 258:C967C981
9. Currin RT, Gores GJ, Thurman RG, et al:
Protection by acidotic pH against anoxic cell
killing in perfused rat liver: evidence for a pH
paradox. FASEB J 1991; 5:207210
10. Nomura F, Aoki M, Forbess JM, et al: Effects
of hypercarbic acidotic reperfusion on recovery of myocardial function after cardioplegic
ischemia in neonatal lambs. Circulation
1994; 90:1132111327
11. Caldwell-Kenkel JC, Currin RT, Coote A, et
al: Reperfusion injury to endothelial cells
after cold storage of rat livers: protection by
mildly acidic pH and lack of protection by
antioxidants. Transpl Int 1995; 8:77 85
12. Moore TM, Khimenko PL, Taylor AE: Restoration of normal pH triggers ischemiareperfusion injury in lung by Na/H exchange activation. Am J Physiol 1995; 269:
H1501H1505
13. Cardenas VJ Jr, Zwischenberger JB, Tao W, et
al: Correction of blood pH attenuates
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. Shibata K, Cregg N, Engelberts D, et al: Hypercapnic acidosis may attenuate acute lung
injury by inhibition of endogenous xanthine
oxidase. Am J Respir Crit Care Med 1998;
158:1578 1584
30. Takeshita K, Suzuki Y, Nishio K, et al: Hypercapnic acidosis attenuates endotoxininduced nuclear factor-[kappa] B activation.
Am J Respir Cell Mol Biol 2003; 29:124 132
31. Adding LC, Agvald P, Persson MG, et al: Regulation of pulmonary nitric oxide by carbon
dioxide is intrinsic to the lung. Acta Physiol
Scand 1999; 167:167174
32. Laffey JG, Engelberts D, Duggan M, et al:
Carbon dioxide attenuates pulmonary impairment resulting from hyperventilation.
Crit Care Med 2003; 31:2634 2640
33. Nuckton TJ, Alonso JA, Kallet RH, et al: Pulmonary dead space fraction as a risk factor
for death in the acute respiratory distress
syndrome. N Engl J Med 2002; 346:
12811286
34. Hough CL, Kallet RH, Ranieri VM, et al:
Intrinsic positive end-expiratory pressure in
Acute Respiratory Distress Syndrome (ARDS)
Network subjects. Crit Care Med 2005; 33:
527532
35. Brower RG, Lanken PN, MacIntyre N, et al:
National Heart, Lung, and Blood Institute
ARDS Clinical Trials Network. Higher versus
lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med 2004; 351:327336