Pulmonary Function in Uremic Patients on Long‐term Hemodialysis

Abstract

Twenty patients with end‐stage renal failure who were on maintenance hemodialysis (HD) underwent pulmonary function testing (PFT) before and shortly after an HD session. On pre‐HD PFT, the mean values of all parameters except residual volume (RV) were in the normal range. Mean RV was high (152.9%), and mean diffusing capacity of the lung for carbon monoxide (DLCO) was high‐normal (110.4%). The pre‐HD static inspiratory (PI_max) and expiratory pressures (PE_max) were much lower than normal (67.4% and 36.3%, respectively). After the HD session, repeat PFT revealed a small increase in expiratory flow rates, and a significant drop in PI_max. There was a strong correlation between PI_max and PE_max (r = 0.567, p < 0.01) at the pre‐ and post‐HD stages, indicating that common mechanism(s) are responsible for impairment of both inspiratory and expiratory muscle strength. The well‐preserved DLCO was thought to be due to the use of biocompatible dialyzer membranes. Chronic vascular congestion might be the other explanation of high DLCO.

• Pulmonary
• Uremic patients
• Hemodialysis
• Renal failure

Introduction

Patients with end‐stage renal disease who are on hemodialysis (HD) are at risk for various pulmonary complications.[1] Some of the most frequent problems are pulmonary edema due to fluid overload, pulmonary calcification and fibrosis secondary to disruption of calcium and phosphorous hemostasis, and pulmonary vascular leukostasis due to activation of the complement cascade by bioincompatible dialyzer membranes.[2], [3], [4] Relatively few studies have investigated alterations of pulmonary function in patients with chronic renal failure (CRF), and some of the results are conflicting. There is no agreement on the parameters related to pulmonary diffusing capacity.[5] The degree of respiratory muscle myopathy and the effects of myopathy on pulmonary mechanics in CRF patients are also not fully understood.

The aim of this study was to assess functional changes in the pulmonary mechanics of CRF patients who have undergone long‐term HD treatment, and to evaluate the acute changes in pulmonary function that occur after an HD session. Relationships between certain laboratory parameters and pulmonary mechanics were also investigated.

Materials and Methods

The study involved 20 consecutive CRF patients who were undergoing HD three times weekly in our institution's Nephrology Department. The demographic characteristics and etiologies of renal failure in this group are shown in Table 1. The exclusion criteria were chronic disease of the lung or chest wall, renal disease affecting the lungs or the respiratory muscles, use of drugs with known detrimental effects on muscle function, such as corticosteroids, current or previous smoking habit and signs of hypervolemia. All patients were ambulatory and in stable clinical condition at the time of the investigation, and all were being examined at regular intervals in the outpatient nephrology clinic. Each individual gave his or her informed consent to participate in the study.

Table 1. Patient Characteristics (Mean ± SD) and the Causes of Chronic Renal Failure

	CRF patients on HD (n = 20)
Age (yrs)	36.7 ± 11.6
Sex (M/F)	12/8
Duration of illness (mo)	86.1 ± 59.4
Duration of HD (mo)	76.3 ± 37.5
Diagnoses
Nephrolithiasis	6
Vesicoureteral reflux	2
Amyloidosis	2
Adult polycystic kidney disease	2
Hypertension	1
Glomerulonephritis	1
Diabetes mellitus	1
Unknown	5

CRF: chronic renal failure; HD: hemodialysis.

As part of the outpatient protocol, each patient had a monthly hemogram done and underwent monthly testing for serum levels of creatinine, blood urea nitrogen, C‐reactive protein, albumin, electrolytes and parathyroid hormone. These same parameters were also assessed in the week prior to the study. In addition, a chest x‐ray of each subject was obtained before the investigation and was interpreted by a pulmonary specialist.

On the day of the study, each individual underwent a complete spirometric study (Sensormedics, Vmax Spectra, Bilthoven, Netherlands) before and shortly after (within 3 hours) an HD session. Dialysis was performed using hemophane membranes at 4–5 hours duration and dialysate containing bicarbonate (30 mEq/L). The mean Kt/V was 1.57 ± 0.2. The blood flow was kept approximately constant at 300 mL/min. The bicarbonate dialysate consisted of 135 mEq/L of sodium, 2.5 mEq/L of potassium, 1.5 mEq/L of magnesium, 3.5 mEq/L of calcium, 106.5 mEq/L of chloride, 8 mEq/L of acetate and 30 mEq/L of bicarbonate.

Total lung capacity (TLC), vital capacity (VC), and residual volume (RV) were measured using the helium dilution technique. Forced expiratory volume in one second (FEV₁), forced vital capacity (FVC), and forced expiratory flow between 25% and 75% of vital capacity (FEF _25–75) were measured using the same apparatus, and the FEV₁/FVC ratio was calculated. The diffusing capacity of the lung for carbon monoxide (DLCO) was measured by the carbon monoxide single‐breath test. Global respiratory muscle strength was assessed by measuring static inspiratory and expiratory pressures (PI_max and PE_max, respectively) according to the method of Black and Hyatt.[6] At least three reproducible tests were carried out for each measurement, and the best result was recorded. Each recorded result was compared to the percent of predicted value for that parameter. A measurement was considered normal when the recorded value was > 80% of the corresponding reference datum.

Statistical Analysis

Statistical analysis was done using the Student's paired t‐test to evaluate differences in the mean values from before to after HD. Relationships between continuous variables were assessed with the Pearson product–moment correlation coefficient (r). The criterion for statistical significance was set at p < 0.05.

Results

As Table 1 indicates, the mean durations of CRF (approximately 7 years) and HD treatment (approximately 6 years) in our patient group were considerably long. The mean volume removed by HD in the sessions on the study day was 2.9 liters (SD ± 0.98).

Table 2 shows the patients' hemogram and blood biochemistry results at the time of the study.

Table 2. The Hemogram and Blood Biochemistry Findings for the Group at the Time of the Study

Hemoglobin (g/dL)	11.9 (1.7)
Hematocrit (%)	38 (0.3)
C‐reactive protein (mg/L)	12.6 (11.6)
Blood urea nitrogen (mg/dL)	87.0 (19.5)
Creatinine (mg/dL)	11.0 (2.8)
Parathormone (ng/)	239.0 (230.3)
Albumin (g/dL)	3.8 (0.3)
Sodium (mEq/L)	134.8 (5.3)
Potassium (mEq/L)	5.4 (0.5)
Calcium (mEq/L)	9.0 (0.8)
Phosphorus (mEq/L)	5.4 (1.6)

Values are expressed as mean (SD).

The chest X‐ray findings and the pulmonary function test (PFT) results before the HD session are presented in Table 3. Although none of the patients had chronic pulmonary conditions such as bronchial asthma or chronic obstructive pulmonary disease, only 10% of the group had normal chest X‐rays. The most common pathologies detected on the films were increased lung markings, areas of hyperlucency in the lung fields, and increased caliber of vessels in the upper zones due to elevated pulmonary venous pressure. The pre‐HD PFT interpretations were normal in 60% of the cases. Restrictive pattern was the most frequent type of disorder. The finding of a high proportion of cases with restrictive pattern combined with low mean serum sodium suggested hypervolemia, which explained most of the abnormalities, observed on the patients' chest X‐rays. Although patients with the signs of hypervolemia were not involved in the study, PFT results together with chest X‐ray findings revealed that in the majority of the patients, pulmonary vascular congestion was present.

Table 3. The Patients' Chest X‐Ray Findings and Pulmonary Function Test Interpretations from Before the Dialysis Session

Chest X‐ray findings	N
Normal	2
Increased bronchovascular markings	5
Hyperlucency	6
Pulmonary venous congestion	4
Unilateral diaphragmatic elevation	1
Bilateral hilar enlargement	2
PFT interpretation
Normal	12
Obstructive pattern	1
Restrictive pattern	5
Mixed pattern	1
Low DLCO	1
High DLCO	2

DLCO: diffusing capacity of the lung for carbon monoxide.

Table 4 shows the findings for the PFT parameters before and after an HD session. Considering the pre‐HD PFT results, all the mean values except RV were in the normal range. Mean RV was elevated at 152.9%. Post‐HD DLCO was apparently unaffected by the patients' high‐normal DLCO prior to the session. After the session, nine patients had lower DLCO, eight had higher DLCO, and three showed no change in this parameter.

Table 4. Comparison of the Findings for the PFT Parameters Before and After an HD Session

Parameters	Predialysis (%)	Postdialysis (%)	p
FVC	83.9 ± 13.1	83.2 ± 10.5	NS
FEV1	86.0 ± 13.8	87.6 ± 11.7	NS
FEF25–75	79.3 ± 20.9	83.1 ± 22.3	NS
TLC	107.8 ± 18.0	105.5 ± 14.2	NS
VC	85.5 ± 13.5	86.2 ± 14.0	NS
RV	152.9 ± 48.2	143.9 ± 34.2	NS
DLCO	110.4 ± 26.2	109.0 ± 23.6	NS
PImax	67.4 ± 20.9	60.9 ± 18.0	< 0.05
PEmax	36.3 ± 13.4	33.7 ± 10.9	NS

NS = not significant.

Values are expressed as percent of predicted value ± SD.

When the pre‐ and post‐HD results were compared using the paired t‐test, the only statistical difference was in PI_max, which was significantly lower after the session (p < 0.05). The mean values for VC, FEV₁ and FEF_25–75 were slightly higher after the session, and all other parameters were slightly lower after the session.

Correlation analysis revealed no relationship between DLCO and duration of dialysis treatment, or between DLCO and duration of disease. As well, none of the biochemical parameters (serum levels of electrolytes, albumin, parathormone, C‐reactive protein) was correlated with DLCO.

There was no correlation between DLCO and volume removed during the HD session, and neither PI_max nor PE_max was correlated with volume removed during HD. There was a strong correlation between PI_max and PE_max (r = 0.567, p < 0.01) both before and after the HD session. This shows that common mechanism(s) are responsible for impairment of both inspiratory and expiratory muscle strength in CRF patients. Biochemical parameters (serum levels of electrolytes, albumin, parathormone, C‐reactive protein) were not correlated with static respiratory muscle strengths.

After the HD session, PI_max values were higher in 6 patients and lower in 14 patients. The PE_max values after HD were higher in 6 patients, lower in 13 patients, and unchanged in 1 patient.

Discussion

The aim of this study was to assess functional changes in the pulmonary mechanics of CRF patients who have undergone long‐term HD treatment, and to evaluate the acute changes in pulmonary function that occur after an HD session. For this purpose, 20 consecutive CRF patients, who were undergoing HD three times weekly in our institution's Nephrology Department, underwent a complete spirometric study (Sensormedics, Vmax Spectra, Bilthoven, Netherlands) before and shortly after (within 3 hours) an HD session.

The CRF patients in this study showed no significant decrease in DLCO from before to after an HD session. Previous investigations have demonstrated below‐normal DLCO in CRF patients on HD.[7], [8], [9] Some authors have explained this on the basis of complement system activation and leukostasis‐related changes in pulmonary vasculature due to the use of bioincompatible dialyzer membranes.[5], [10] One study showed that patients who had been on HD for more than 1 year exhibited more severely reduced DLCO than patients with shorter dialysis duration.[5] Although the mean duration of HD in our study group was long (approximately 6 years), our 20 patients showed good preservation of DLCO. This might be explained by the fact that we used cellulosynthetic hemophane membranes, which are considered to activate the complement system weakly compared to bioincompatible membranes. At our institution, we have used cellulosynthetic membranes for more than 5 years. Due to the uneven distribution of patients who had been on HD for short (less than 1 year) versus long periods of time, we were unable to compare findings in such groups in our study.

Chan et al.[11] studied pulmonary function in a group of CRF patients on HD who underwent renal transplantation. The patients' DLCO and RV were in the high range of predicted values (115.7% and 157.8%, respectively) before transplantation, but the values dropped steadily after transplantation and normalized at 6 months postsurgery. The authors attributed the high DLCO and RV values pretransplantation to chronic vascular congestion. Their results are similar to the DLCO and RV values we recorded in our study. Our patients' chest X‐ray findings, the predominant restrictive pattern (n = 5) of pulmonary function in the group, and low mean serum sodium level also suggest hypervolemia. In addition to the above‐mentioned use of more biocompatible membranes, a chronic congestive state might have contributed to the high DLCO and RV values we recorded. However, we noted only minor reductions in RV and DLCO from pre‐ to postdialysis, even though quite amount of fluid was removed (2.90 L ± 0.98) and this does not support the hypothesis of Chan et al. In our study, hypervolemia did not seem to be, at least, the predominant mechanism by which DLCO was preserved.

In our investigation of CRF patients on HD, pre‐HD testing showed that the majority had well‐preserved lung volumes and expiratory flow rates. Only one patient exhibited a mild obstructive pattern (FEV₁/FVC = 69%), and one other individual showed a mixed pattern. Reports have stated that, in CRF patients on HD, lung volumes and flow rates are normal unless there is chronic pulmonary disease or chest wall disease.[5], [9] The elevated RV in our patient group may be explained by impaired deflation of the lungs due to vascular congestion.

Comparison of the PFT results from before and after the session revealed a small increase in expiratory flow rates, which reflects improved patency of both smaller (FEF_25–75) and larger (FEV₁) airways. We believe that this increase was related to reduction of mucosal edema in the airways after volume reduction.

Our initial spirometric tests revealed that the patient's mean PI_max and mean PE_max were both well below normal (67.4% and 36.3% of predicted value, respectively). Reports in the literature state that respiratory muscle strength is impaired in CRF patients on HD.[12], [13], [14] Bark et al. discussed the possible mechanisms for this impairment,[12] and highlighted abnormal vitamin D metabolism, excess parathyroid hormone, uremic toxins and malnutrition as possible factors involved in both skeletal and respiratory muscle weakness. Many of these factors also apply to our patient group. Although Bark et al. found a positive correlation between phosphorus level and respiratory muscle strength; our analysis indicated no relationships between biochemical parameters (including electrolyte levels) and muscle strength. Hyperphosphatemia was common (14/20 patients) in our study group, but our analysis failed to show preservation of muscle function. Gomez et al. described a uremic dialysis patient who had severe secondary hyperparathyroidism and exhibited impaired respiratory muscle strength.[15] In that case, the authors observed a marked improvement in respiratory muscle strength after subtotal parathyroidectomy. We failed to show any difference in both expiratory and inspiratory muscle strength between the patients with normal blood PTH levels and patients with secondary hyperparathyroidism.

We also found a significant correlation between PI_max and PE_max (r = 0.567, p < 0.01) on pre‐HD session PFT, indicating similar involvement of the inspiratory and expiratory muscle groups. The recorded PE_max levels were much lower than the PI_max values, and we believe this is because the reference values we used were those provided by Black and Hyatt.[6] Previous reports have stated that the Black and Hyatt equations are not valid for some populations.[16], [17] Ordiales et al. documented low PE_max values (69.7%) in 100 healthy subjects from Spain.[16] We cannot make a definitive statement about this because there was no healthy control group in our study, but our opinion is that the Black and Hyatt equations are responsible for the extremely low PE_max values in our patient group.

We found only one study in the literature that investigated the acute effects of HD on respiratory muscle strength. In that report, Weiner et al. documented increased PI_max values after a dialysis session in 13 of 21 patients.[13] They attributed this improvement to dialysis‐induced changes in biochemical parameters. The majority of our patients showed decreased inspiratory muscle strength (70% of patients) and expiratory muscle strength (65% of patients) after a dialysis session. In our opinion, dialysis‐induced improvement in biochemical parameters due to removal of uremic toxins and correction of electrolyte levels do not result in improved inspiratory muscle performance. The catabolic effects of HD may be responsible for this impairment. In uremic patients, HD leads carnitine depletion due to loss through dialytic membranes. As a complication of this metabolic deficit, impairment of exercise capacity, muscle symptoms were seen in this patient group.[18], [19] To evaluate the effects of carnitine deficiency on respiratory muscle function, studies that concern respiratory muscle strength after carnitine replacement are needed.

In conclusion, our results indicate that flow rates and lung volumes are well preserved in CRF patients who have been on long‐term maintenance HD. We also found that DLCO was close to normal in our patient group, and believe this is largely related to the use of more biocompatible dialyzer membranes. Testing revealed significant impairment of inspiratory and expiratory muscle strength. We observed significantly lower PI_max values after the HD session, and we attribute this drop to the catabolic effects of dialysis.