Serum and Lavage Lactate Dehydrogenase Isoenzymes in Pulmonary Alveolar Proteinosis1- 3
ROBERT M. HOFFMAN and ROBERT M. ROGERS
Introduction
Pulmonary alveolar proteinosis (PAP) is a disease of unknown etiology characterized by the accumulation of lipoproteinaceous material in the pulmonary alveolar spaces of afflicted individuals (1-3). Treatment with therapeutic whole lung lavagecan bring about dramatic improvement in the clinical and physiologic status of patients with this disease (4). The association of PAP with an elevated serum level of lactate dehydrogenase (LDH) wasfirst noted by Ramirez-R, and colleagues in 1963 (5). In several subsequent reports, the LDH concentration also seemed to correlate with disease activity in that it decreased into the normal range when a remission was brought about by treatment with lavage (6-10). In a more recent series, 14 of 18 patients with PAP had an elevated serum LDH concentration, whereas only five of 17 patients with other types of diffuse lung disease had high LDH concentrations (11, . 12). It has also been noted that the serum LDH isoenzyme distribution in PAP was normal, although no data were given in this report (10). Despite the fact that elevated serum LDH has been found in many cases of PAP, it is not a universal finding. We therefore decided to try to refine these observations in several ways. First, we identified the total serum LDH concentration and isoenzyme pattern in patients with PAP before whole lung lavage. Second, wecompared total serum LDH concentration and isoenzyme pattern before and after treatment with lavage. Finally, we looked at the total LDH concentration and isoenzyme pattern in the lavage fluid itself and attempted to correlate these levels with those found in the serum. Methods All patients in this study were referred to Presbyterian-University Hospital, Pittsburgh, Pennsylvania, for evaluation and treatment of PAP. This diagnosis had been made by open lung biopsy in 11 patients and by transbronchial lung biopsyin two patients. Ther-
42
SUMMARY Pulmonary alveolar protelnosls (PAP)Is a rare disease characterized by the accumulation of llpoprotelnaceous material In the alveolar space. Serum lactate dehydrogenase (LDH) has been noted to be elevated In patients with PAP In previous studies. We sought to extend this observation In a series of patients with PAP by looking at total serum LDH concentrations and LDH Isoenzyme fractions measured before and after whole lung lavage. Total LDH and LDH Isoenzymes were also determined In the lavage effluent. Total serum LDH was elevated before lavage In 10 of 16 patients. Prelavage serum LDH and prelavage alveolar-arterial O2 gradient showed a significant correlation (r = 0.62, P < 0.05). A decrease In serum LDH was found after lavage In all patients In whom postlavage data was available (paired t test, p < 0.01, n = 11), although the magnitude of this decrease varied considerably. The Isoenzyme pattern before lavage was Isomorphic, and this pattern was unchanged after whole lung lavage. This was In marked contrast to the LDH Isoenzyme pattern observed In the lavage effluent, which showed a lower percent LDH1 and LDH2 and a higher percent LDH3, LDH4, and LDH5 when compared with the corresponding prelavage Isoenzyme percentages for serum (unpaired t test, p < 0.001). There was no correlation between the total serum LDH concentration and the total lavage LDH concentration. These data confirm that elevated serum LDH Is a common finding In PAP. Furthermore, the LDH elevation found consistently In the alveolar fluid points to this as the source of the serum LDH. In addition, this is the first time a sufficient number of patients has been available to statistically demonstrate a decrease In serum LDH levels after treatment with whole lung lavage. The large quantities of LDH found In lavage effluent and the Isoenzyme pattern of the lavage LDH suggest that pulmonary ceil death occurs In PAP to a considerable extent, but the specific cells that contribute to the lavage LDH and the Impact of Intraalveolar LDH on serum LDH require further study. AM REV RESPIR DIS 1991; 143:42-46
apeutic whole lung lavage was performed under general anesthesia as has been previously described (2). It is our practice to lavage the lung that is more severely involved first and then to lavage the contralateral lung exactly one week later. All patients in this series were lavaged in this fashion except for one patient in which the second lung was lavaged 6 wk after the first lung (Patient 8). 1\vo patients had both lungs lavaged in the standard fashion on several occasions: Patient 2, who had both lungs lavaged on three separate occasions at yearly intervals and Patient 3, who had both lungs lavaged on two separate occasions 18months apart. Arterial blood gases were drawn on all patients before lavage of the first lung. Serum was drawn for total LDH and LDH isoenzyme determination immediately before the lavage of the first lung. Additional serum was obtained for total LDH and LDH isoenzymes after the lavageof the contralateral lung and before the patient's discharge from the hospital. This interval was usually 3 days but ranged from 2 to 8 days. Postlavage values were not available in all instances due to clerical errors (2), loss of specimens in transit to the reference laboratory (2), and inadvertent freezing of a specimen, which makes the LDH assay inaccurate. Because several different
laboratories were used to determine the serum LDH concentrations, the normal ranges tended to vary. Therefore, the results are expressed as the percentage change above (given as a positive percent) or below (given as a negative percent) the upper normal limit, as previously described (11, 12). Lavageeffluent was assessed for total LDH and LDH isoenzymes in the following manner. A lOO-ml aliquot from the first 1,500 ml of effluent from the patient's whole lung la-
(Received in original form March 29, 1990 and in revised form August 14, 1990) 1 From the Division of Pulmonary and Critical Care Medicine. Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. 2 Supported by Clinical Investigator Award No. HL-01367 from the National Heart. Lung, and Blood Institute of the National Institutes of Health, from the Presbyterian-University Hospital, and from the Christmas Seal League of Western Pennsylvania. 3 Correspondence and requests for reprints should be addressed to Robert M. Hoffman, M.D., Pulmonary Division. Department of Medicine,440 Scaife Hall, 3550 Terrace Street. Pittsburgh, PA 15261.
43
LDH IN ALVEOLAR PROTEINOSIS
resis on cellulose acetate was used for LDH isoenzyme determination. Data are expressed as the mean ± So. Serum LDH and isoenzyme values before and after the whole lung lavage were compared using the paired t test. Linear regression was used to analyze for correlation betweenserum LDH and alveolar-arterial O 2 gradient and between serum and lavage LDH levels (13).
TABLE 1 TOTAL SERUM LDH CONCENTRATION* AND ALVEOLAR-ARTERIAL O2 GRADIENT IN PATIENTS UNDERGOING LAVAGE
Patient No. 1 2a 2b 2c 3a 2b 4 5
6 7 8 9 10 11 12 13 Mean SO n
LDH before Lavage
LDH after Lavaget
Alveolar-arterial O 2 Gradient before Lavage (mmHg)
-37.6 69.6 35.3 -3.6 -13.6 -17.2 99.2 13.2 29.2 -24 31.6 26.5 -2.0 21.2 97.7 35.9
-47.2 17.1 9.4 NA -18 NA 97.6 -5.3 21.6 -28 NA 6.8 NA NA 27.6 -1.2
NAt 45.0 39.5 48.3 21.2 31.9 NA 47.2 40.7 23.3 NA 41.3 43.9 40.7 NA 61.3
22.6 40.2 16
7.3 37.4 11
40.4 10.9 12
Results
Definition of abbreviations: LDH = lactate dehydrogenase; NA = not available. * Results are expressed as the percentage change above (given as a positive percent) or below (given as a negative percent) the upper normal limit. t For patients with LDH levels measured before and after lavage (n = 11). there was a statistically significant difference (paired t test, p < 0.01). :j:Alveolar-arterial O2 gradients before lavage were not available on four patients with severe hypoxemia due to the inability to safely obtain an arterial blood gas on room air. Serum LDH was not available on five patients after lavage due to the reasons cited in MeTHODS.
vage was centrifuged at 400 x g, and total LDH concentration and LDH isoenzyme fractions weredetermined on the supernatant. The rate of reduction of nicotinamide ade-
nine dinucleotide (NAD) to NADH in the presence of L-Iactate, as reflected by the rate of absorbance increase at 340 nm, was used to quantitate total LDH activity. Electropho-
TABLE 2 SERUM LDH ISOENZYMES* IN PATIENTS UNDERGOING LAVAGE Patient No.
LDH1
LDH2
------
LDH4
LDH3
LDH5
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Pre
Post
13 NAt NA NA 23 NA 21 22 17 24 NA 23 NA NA 23 NA
33 36 38 40 34 33 36 32 35 29 35 35 32 29 NA NA
41 NA NA NA 32 NA 32 33 30
22 26 24 25 20 22 22 24 27 23 27 26 23 34 NA NA
29 NA NA NA 20 NA 24 24 23 21 NA 22 NA NA 21 NA
8 12 10 9 9 10 10 12 10 12 13 11 10 9 NA NA
9 NA NA NA 10 NA 12 11 11 10 NA 10 NA NA
10 9 9
9 10 11 12 13
27 17 19 20 25 22 25 19 16 21 15 21 21 16 NA NA
12 13 7 13 12 15 10 7 14 12 NA NA
8 NA NA NA 15 NA 11 10 19 11 NA 9 NA NA 7 NA
Mean SO n
20 4 14
21 4
34 3 14
25 3 14
23 3
10 2 14
11 3 14
11 4 8
1 2a 2b 2c 3a 3b 4 5 6 7
8
Normal
8 17-27
34 NA 36 NA NA 41 NA 35 4
8 28-38
8 19-27
8 NA 10 1
8 5-16
6
5-16
For definition of abbreviations, see table 1. * Results are expressed as the percent of total LDH. t Serum LDH was not available on five patients after lavage due to the reasons cited in MeTHODS. In addition, isoenzymes were not performed on two patients before lavage and on three patients after lavage.
The total serum LDH concentration before lavage being performed on the first lung and after both lungs had been lavaged is summarized in table 1. Also included in the table is the measured prelavagealveolar-arterial O 2 gradient on room air for each patient in which it was able to be determined. Ten of 16patients had an elevated serum LDH concentration (value greater than zero) before whole lung lavage. In each of the 11 instances in which serum LDH levels were obtained before and after lavage of both lungs, there was a decrease in the serum LDH concentration after lavage, although the magnitude of this decrease varied. This decrease was statistically significant (paired t test, p < 0.01). There was a statistically significant correlation between the prelavage total serum LDH concentration and alveolar-arterial O 2 gradient (r = 0.62, p < 0.05). The serum LDH isoenzyme pattern before lavage being performed on the first lung and after completion of the lavage on the second lung is shown in table 2. Results are expressed as the percent of total LDH concentration. The mean percent value fell within the normal range for each isoenzyme before and after both lungs werelavaged. Paired isoenzyme data before and after lavage were available for seven of the 16total LDH concentrations listed in table 1. For these patients, no statistically significant difference was noted between the isoenzyme percentages before and after lavage (paired t test). The concentration of total LDH and the LDH isoenzyme pattern found in the PAP lavage supernatant collected from 22 whole lung lavages is shown in table 3. The mean value was 276 lUlL (range, 56 to 797 lUlL). There was no correlation between the prelavage total serum LDH concentration and lavage total LDH concentration (p = 0.8, r = 0.1). In the lavage fluid, there was a higher percent of LDHI and LDH2 and a lower percent value for LDH3, LDH4, and LDH5 when compared with the corresponding prelavage isoenzyme percentages for serum (unpaired t test, P
ROBERT M. HOFFMAN and ROBERT M. ROGERS
Introduction
Pulmonary alveolar proteinosis (PAP) is a disease of unknown etiology characterized by the accumulation of lipoproteinaceous material in the pulmonary alveolar spaces of afflicted individuals (1-3). Treatment with therapeutic whole lung lavagecan bring about dramatic improvement in the clinical and physiologic status of patients with this disease (4). The association of PAP with an elevated serum level of lactate dehydrogenase (LDH) wasfirst noted by Ramirez-R, and colleagues in 1963 (5). In several subsequent reports, the LDH concentration also seemed to correlate with disease activity in that it decreased into the normal range when a remission was brought about by treatment with lavage (6-10). In a more recent series, 14 of 18 patients with PAP had an elevated serum LDH concentration, whereas only five of 17 patients with other types of diffuse lung disease had high LDH concentrations (11, . 12). It has also been noted that the serum LDH isoenzyme distribution in PAP was normal, although no data were given in this report (10). Despite the fact that elevated serum LDH has been found in many cases of PAP, it is not a universal finding. We therefore decided to try to refine these observations in several ways. First, we identified the total serum LDH concentration and isoenzyme pattern in patients with PAP before whole lung lavage. Second, wecompared total serum LDH concentration and isoenzyme pattern before and after treatment with lavage. Finally, we looked at the total LDH concentration and isoenzyme pattern in the lavage fluid itself and attempted to correlate these levels with those found in the serum. Methods All patients in this study were referred to Presbyterian-University Hospital, Pittsburgh, Pennsylvania, for evaluation and treatment of PAP. This diagnosis had been made by open lung biopsy in 11 patients and by transbronchial lung biopsyin two patients. Ther-
42
SUMMARY Pulmonary alveolar protelnosls (PAP)Is a rare disease characterized by the accumulation of llpoprotelnaceous material In the alveolar space. Serum lactate dehydrogenase (LDH) has been noted to be elevated In patients with PAP In previous studies. We sought to extend this observation In a series of patients with PAP by looking at total serum LDH concentrations and LDH Isoenzyme fractions measured before and after whole lung lavage. Total LDH and LDH Isoenzymes were also determined In the lavage effluent. Total serum LDH was elevated before lavage In 10 of 16 patients. Prelavage serum LDH and prelavage alveolar-arterial O2 gradient showed a significant correlation (r = 0.62, P < 0.05). A decrease In serum LDH was found after lavage In all patients In whom postlavage data was available (paired t test, p < 0.01, n = 11), although the magnitude of this decrease varied considerably. The Isoenzyme pattern before lavage was Isomorphic, and this pattern was unchanged after whole lung lavage. This was In marked contrast to the LDH Isoenzyme pattern observed In the lavage effluent, which showed a lower percent LDH1 and LDH2 and a higher percent LDH3, LDH4, and LDH5 when compared with the corresponding prelavage Isoenzyme percentages for serum (unpaired t test, p < 0.001). There was no correlation between the total serum LDH concentration and the total lavage LDH concentration. These data confirm that elevated serum LDH Is a common finding In PAP. Furthermore, the LDH elevation found consistently In the alveolar fluid points to this as the source of the serum LDH. In addition, this is the first time a sufficient number of patients has been available to statistically demonstrate a decrease In serum LDH levels after treatment with whole lung lavage. The large quantities of LDH found In lavage effluent and the Isoenzyme pattern of the lavage LDH suggest that pulmonary ceil death occurs In PAP to a considerable extent, but the specific cells that contribute to the lavage LDH and the Impact of Intraalveolar LDH on serum LDH require further study. AM REV RESPIR DIS 1991; 143:42-46
apeutic whole lung lavage was performed under general anesthesia as has been previously described (2). It is our practice to lavage the lung that is more severely involved first and then to lavage the contralateral lung exactly one week later. All patients in this series were lavaged in this fashion except for one patient in which the second lung was lavaged 6 wk after the first lung (Patient 8). 1\vo patients had both lungs lavaged in the standard fashion on several occasions: Patient 2, who had both lungs lavaged on three separate occasions at yearly intervals and Patient 3, who had both lungs lavaged on two separate occasions 18months apart. Arterial blood gases were drawn on all patients before lavage of the first lung. Serum was drawn for total LDH and LDH isoenzyme determination immediately before the lavage of the first lung. Additional serum was obtained for total LDH and LDH isoenzymes after the lavageof the contralateral lung and before the patient's discharge from the hospital. This interval was usually 3 days but ranged from 2 to 8 days. Postlavage values were not available in all instances due to clerical errors (2), loss of specimens in transit to the reference laboratory (2), and inadvertent freezing of a specimen, which makes the LDH assay inaccurate. Because several different
laboratories were used to determine the serum LDH concentrations, the normal ranges tended to vary. Therefore, the results are expressed as the percentage change above (given as a positive percent) or below (given as a negative percent) the upper normal limit, as previously described (11, 12). Lavageeffluent was assessed for total LDH and LDH isoenzymes in the following manner. A lOO-ml aliquot from the first 1,500 ml of effluent from the patient's whole lung la-
(Received in original form March 29, 1990 and in revised form August 14, 1990) 1 From the Division of Pulmonary and Critical Care Medicine. Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. 2 Supported by Clinical Investigator Award No. HL-01367 from the National Heart. Lung, and Blood Institute of the National Institutes of Health, from the Presbyterian-University Hospital, and from the Christmas Seal League of Western Pennsylvania. 3 Correspondence and requests for reprints should be addressed to Robert M. Hoffman, M.D., Pulmonary Division. Department of Medicine,440 Scaife Hall, 3550 Terrace Street. Pittsburgh, PA 15261.
43
LDH IN ALVEOLAR PROTEINOSIS
resis on cellulose acetate was used for LDH isoenzyme determination. Data are expressed as the mean ± So. Serum LDH and isoenzyme values before and after the whole lung lavage were compared using the paired t test. Linear regression was used to analyze for correlation betweenserum LDH and alveolar-arterial O 2 gradient and between serum and lavage LDH levels (13).
TABLE 1 TOTAL SERUM LDH CONCENTRATION* AND ALVEOLAR-ARTERIAL O2 GRADIENT IN PATIENTS UNDERGOING LAVAGE
Patient No. 1 2a 2b 2c 3a 2b 4 5
6 7 8 9 10 11 12 13 Mean SO n
LDH before Lavage
LDH after Lavaget
Alveolar-arterial O 2 Gradient before Lavage (mmHg)
-37.6 69.6 35.3 -3.6 -13.6 -17.2 99.2 13.2 29.2 -24 31.6 26.5 -2.0 21.2 97.7 35.9
-47.2 17.1 9.4 NA -18 NA 97.6 -5.3 21.6 -28 NA 6.8 NA NA 27.6 -1.2
NAt 45.0 39.5 48.3 21.2 31.9 NA 47.2 40.7 23.3 NA 41.3 43.9 40.7 NA 61.3
22.6 40.2 16
7.3 37.4 11
40.4 10.9 12
Results
Definition of abbreviations: LDH = lactate dehydrogenase; NA = not available. * Results are expressed as the percentage change above (given as a positive percent) or below (given as a negative percent) the upper normal limit. t For patients with LDH levels measured before and after lavage (n = 11). there was a statistically significant difference (paired t test, p < 0.01). :j:Alveolar-arterial O2 gradients before lavage were not available on four patients with severe hypoxemia due to the inability to safely obtain an arterial blood gas on room air. Serum LDH was not available on five patients after lavage due to the reasons cited in MeTHODS.
vage was centrifuged at 400 x g, and total LDH concentration and LDH isoenzyme fractions weredetermined on the supernatant. The rate of reduction of nicotinamide ade-
nine dinucleotide (NAD) to NADH in the presence of L-Iactate, as reflected by the rate of absorbance increase at 340 nm, was used to quantitate total LDH activity. Electropho-
TABLE 2 SERUM LDH ISOENZYMES* IN PATIENTS UNDERGOING LAVAGE Patient No.
LDH1
LDH2
------
LDH4
LDH3
LDH5
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Pre
Post
13 NAt NA NA 23 NA 21 22 17 24 NA 23 NA NA 23 NA
33 36 38 40 34 33 36 32 35 29 35 35 32 29 NA NA
41 NA NA NA 32 NA 32 33 30
22 26 24 25 20 22 22 24 27 23 27 26 23 34 NA NA
29 NA NA NA 20 NA 24 24 23 21 NA 22 NA NA 21 NA
8 12 10 9 9 10 10 12 10 12 13 11 10 9 NA NA
9 NA NA NA 10 NA 12 11 11 10 NA 10 NA NA
10 9 9
9 10 11 12 13
27 17 19 20 25 22 25 19 16 21 15 21 21 16 NA NA
12 13 7 13 12 15 10 7 14 12 NA NA
8 NA NA NA 15 NA 11 10 19 11 NA 9 NA NA 7 NA
Mean SO n
20 4 14
21 4
34 3 14
25 3 14
23 3
10 2 14
11 3 14
11 4 8
1 2a 2b 2c 3a 3b 4 5 6 7
8
Normal
8 17-27
34 NA 36 NA NA 41 NA 35 4
8 28-38
8 19-27
8 NA 10 1
8 5-16
6
5-16
For definition of abbreviations, see table 1. * Results are expressed as the percent of total LDH. t Serum LDH was not available on five patients after lavage due to the reasons cited in MeTHODS. In addition, isoenzymes were not performed on two patients before lavage and on three patients after lavage.
The total serum LDH concentration before lavage being performed on the first lung and after both lungs had been lavaged is summarized in table 1. Also included in the table is the measured prelavagealveolar-arterial O 2 gradient on room air for each patient in which it was able to be determined. Ten of 16patients had an elevated serum LDH concentration (value greater than zero) before whole lung lavage. In each of the 11 instances in which serum LDH levels were obtained before and after lavage of both lungs, there was a decrease in the serum LDH concentration after lavage, although the magnitude of this decrease varied. This decrease was statistically significant (paired t test, p < 0.01). There was a statistically significant correlation between the prelavage total serum LDH concentration and alveolar-arterial O 2 gradient (r = 0.62, p < 0.05). The serum LDH isoenzyme pattern before lavage being performed on the first lung and after completion of the lavage on the second lung is shown in table 2. Results are expressed as the percent of total LDH concentration. The mean percent value fell within the normal range for each isoenzyme before and after both lungs werelavaged. Paired isoenzyme data before and after lavage were available for seven of the 16total LDH concentrations listed in table 1. For these patients, no statistically significant difference was noted between the isoenzyme percentages before and after lavage (paired t test). The concentration of total LDH and the LDH isoenzyme pattern found in the PAP lavage supernatant collected from 22 whole lung lavages is shown in table 3. The mean value was 276 lUlL (range, 56 to 797 lUlL). There was no correlation between the prelavage total serum LDH concentration and lavage total LDH concentration (p = 0.8, r = 0.1). In the lavage fluid, there was a higher percent of LDHI and LDH2 and a lower percent value for LDH3, LDH4, and LDH5 when compared with the corresponding prelavage isoenzyme percentages for serum (unpaired t test, P
