© 1991 S. Karger AG. Basel 0025-7931/ 9 1/0582-0072 S 2.75/0
Respiration 1991;58:72-76
Bronchoalveolar Lavage: Comparison of Three Commonly Used Procedures1* M .
Pirozynskia , P. Sliwinskih. L. Radwanc, J. Zielinskih
Departmens of * Bronchology and h Respiratory Medicine, and c Lung Function Unit, Institute of Tuberculosis and Lung Diseases, Warsaw, Poland
Key Words. Bronchoalveolar lavage • Technical aspects • Side effects
Introduction Although bronchoalveolar lavage has been widely used in modern pulmonology for over 10 years, and has been documented to be a valuable tool in moni toring disease course [1-3], in research (for sampling of cells and soluble particles from the alveoli) and di 1 Results of this study were presented in Freiburg during the 8th Societas Europea Pneumonologica Meeting.
agnosis (especially in immunocompromised patients) the obtained results can hardly be compared between centers mainly due to a lack of standardization of the lavage technique [3]. Instilled lavage volumes in the past ranged from 100 to 500 ml [1, 3, 4], Smaller vol umes are said to bear the risk that bronchial washings dominate the overall view [5, 6]. Larger volumes, al though are more representative of the alveolar spaces, may induce dangerous side effects, especially in highrisk patients [7, 8]. The aim of this study was to com-
Downloaded by: King's College London 137.73.144.138 - 11/27/2017 12:51:11 PM
Abstract. We have studied three different lavage procedures (100, 200 and 300 ml) in patients with pulmo nary sarcoidosis (stage I). The effect of bronchoalveolar lavage (BAL) on cell yield, lavage fluid recovery, dwelling time, lavage-induced arterial oxygen desaturation and occurrence of side effects was analyzed. The patients did not differ significantly in prelavage lung function and blood gas parameters. The lowest BAL re turn was seen in the 300-ml lavage procedure (49.5%), while the medium yielded over 70%. The lowest cell yield was seen in the BAL 100 g r o u p (10.4 x 106); the highest in the BAL 300 (19.4 x 106), but the latter did not differ significantly from BAL 200 (18.4 x 106). Dwelling time of the fluid differed only slightly between the small and middle volume lavage (average 3.2 vs. 3.9 min p < 0.01), but was significantly lower from the average dwelling time in the BAL 300 group (9.8 min, p < 0.001). Arterial oxygen desaturation was lowest in the BAL 100 and most pronounced in the large-volume lavage. Side effects were seen in all but I patient undergoing BAL 300. Cough was the most often reported side effect (9 patients); fever was observed in 6 patients, dyspnea in 4 (all undergoing large-volume lavage). Considering our results we do not think that it is justifiable to increase the volume of instilled fluid above 200 ml, because this may lead to serious side effects without increasing benefits. Using lower than 200 ml volumes decrease diagnostic yield although the risk of developing side effects is much lower.
Bronchoalveolar Lavage: Comparison of Three Commonly Used Procedures
pare the effect of the three most commonly used lavage techniques on bronchoalveolar lavage (BAL) return, arterial oxygen saturation, dwelling time and occurrence of side effects, as well as to propose a rea sonably safe lavage procedure, at the same time pro viding wide information.
Materials and Methods Sliuly Group 33 consecutive patients with stage I pulmonary sarcoidosis were included in the study. All were nonsmokers. They did not differ significantly in selected lung function and blood gas parameters (tables I, 2). They were allocated to three different lavage groups. In order to evaluate the role of fiberoptic bronchoscopy on the analysis of the effect o f BAL upon the studied parameters, a control group was made up of 13 consecutive patients undergoing routine diagnostic endoscopy due to suspected pulmonary malignancy.
73
Table 1. General characteristics of studied patients Group
Mean (±SD )age years'
Patients n
Sex ratio F:M
Control BAL 100 BAL 200 BAL 300
49.5 ± 4.5 32.4 ±2.4 41.1 ±2.4 41.0 ±4.4
13 8 18 7
1:12 3:4 1:1 1:3
1 All differences statistically significant (p < 0 .0 0 l), except between BAL 200 and 300 (NS).
Su.OV - SlcO?At m,n ~ s o f -------* 100
a s „o 2--------
where AS„.02 = desaturation rate in percent; S,cO°2 *• prelavage values o f arterial oxygen saturation: S,cO: BA1 m"' = lowest arterial
oxygen saturation reading during the lavage procedure. Side Effects. Each patient was asked to record all observed ad verse effects (i.e. cough, dyspnea, chills, myalgias, etc.) during the subsequent 24 h after the lavage procedure. Medical records were checked for the presence of fever (above 37.5°C)and other observa tions recorded by physicians during the 24 h after the lavage proce dure. Chest radiograms taken 24 h after BAL were reviewed for possible changes. Each patient's Slc0 2 was continuously monitored for 2 h after bronchoalveolar lavage. Additionally a recording of S(c0 2 24 h after BAL was carried out. These readings were related to prelavage values and expressed in percent according to the follow ing formula:
As„o-;jh =
S.cO? - S...O;4h
S„0:
-x 100
where AS,c0 224 h = desaturation rate 24 h following BAL or fiber optic bronchoscopy in percent: S,cO°2 = prelavage or prebronchoscopic values of S,cO:: S^.O:241' = Slt0 2 recorded 24 h after BAL or fiberoptic bronchoscopy. Statistical analysis was carried out using the t test.
Results There were no gross differences in lung function parameters and Pa02 in the analyzed patients that un derwent lavage. The BAL 300 group was the oldest of the lavaged patients, while the control group was the oldest of all (table 1). The poorest BAL return was noted in the large-volume lavage group (BAL 300), the highest in the medium lavage (BAL 200) - exceed ing 70% of the infused saline (differences statistically significant, p < 0.001; table 3).
Downloaded by: King's College London 137.73.144.138 - 11/27/2017 12:51:11 PM
Methods Premedication was not used in any patient. Only local anesthe sia was used (Xylocaine 2% - Astra). The maximal dose was ad justed individually but never exceeded 200 mg. Fiberoptic bron choscopy was performed using the transoral approach, with the pa tient in the supine position. Supplemental oxygen was not given in all evaluated patients. In all cases of diagnostic bronchoscopy for ceps biopsy was carried out. Olympus BF I TR endoscopes were used. Bronchoalveolar Lavage. Following spraying of the nasopha rynx, glottis, and larynx with 2% Xylocaine solution, the endo scope was inserted into the trachea. Immediately a 5-ml bolus of 2% Xylocaine was instilled to abolish cough reflexes. An additional 2 ml was always instilled into the orifice of the middle lobe bron chus. The tip of the endoscope was then advanced and wedged in a subsegmental bronchus (RB4 or RB5). in a position allowing visu alization of the distal airway. Two lavage procedures differed only in the volume of instilled saline. The BAL 100 (small volume) la vage consisted of introducing 5 portions of 20 ml each. The fluid was sucked back using the same syringe after each portion. The pressure was controlled to permit sufficient suction without collap sing the/bronchial walls. BAL 200 (middle volume) lavage was per formed similarly, 10 aliquots of 20 ml each were instilled. A differ ent technique was used for the BAL 300 (large volume) lavage. Saline was instilled in three 100-ml portions and suctioned back (immediately after each portion) into a sterile 500-ml side arm flask adapted to function as a suction trap. Electric suction was used, pressure being regulated by occlusion of the suction part of the en doscope, to prevent closure of the distal bronchial walls. The in stilled saline was always warmed up to the temperature of 37°C. BAL specimens were processed according to the method described by Hunninghake et al. [9]. BAL return was calculated in relation to the infused volume (in percent). Percutaneous Continuous Pulse Oximetry. An Ohmeda Biox 3700 continuous pulse oximeter was used. Continuous pulse oxime try was performed as described earlier by us [10]. Arterial oxygen desaturation was calculated from oximetric readings in relation to the prclavage values, according to the following formula:
74
Pirofcynski/Sliwihski/Radwan/Zielihski
Table 2. Selected lung function and blood gas parameters in studied patients Group
VC%PRED %
FEV,%PRED %
RV%TLC %
Raw cm H:0 / l/s
Pst/TLC cm H ,0
Cst/TGV
PaO, mm Hg
Control BAL 100 BAL 200 BAL 300
86.0 ±5.9 92.3 ± 5.6 97.3 ± 5.0 96.6 ±4.9
78.1 ±8.2 98.9 ±5.5 101.7 ±9.5 100.6 ±6.8
_
_
_
_
25.6 ± 3.1 24.2 ± 3.3 23.5 ±2.9
1.5 ±0.2 1.2 ±0.2 1.2 ±0.3
29.5 ±4.6 24.4 ±3.5 25.8 ±2.9
0.54 ±0.09 0.51 ±0.09 0.52 ± 0.05
78.7 ±2.5 84.8 ± 2.4 86.4 ±3.5 87.1 ±2.8
All differences between lavaged patients nonsignificant, differences between control and lavaged groups statistically significant ( p < 0.001).
Table 4. Side effects observed in control and lavage group Group
Group
BAL return' %
AS,cO; %
Cell yield' x I06
_
4.2 ±0.7 _ 6.3 ±0.9 10.3 ± 1.7 8.4 ± 1.2 18.4 ± 1.2 11.5 ± 1.7 19.4 ± 1.9
Dwelling time4, min
AS,c0 24h %
Side effects type
patients, n
Control
hemoptysis fever
3 2
0.01
BAL 100
cough
1
0.015
1 All differences statistically significant (p
Respiration 1991;58:72-76
Bronchoalveolar Lavage: Comparison of Three Commonly Used Procedures1* M .
Pirozynskia , P. Sliwinskih. L. Radwanc, J. Zielinskih
Departmens of * Bronchology and h Respiratory Medicine, and c Lung Function Unit, Institute of Tuberculosis and Lung Diseases, Warsaw, Poland
Key Words. Bronchoalveolar lavage • Technical aspects • Side effects
Introduction Although bronchoalveolar lavage has been widely used in modern pulmonology for over 10 years, and has been documented to be a valuable tool in moni toring disease course [1-3], in research (for sampling of cells and soluble particles from the alveoli) and di 1 Results of this study were presented in Freiburg during the 8th Societas Europea Pneumonologica Meeting.
agnosis (especially in immunocompromised patients) the obtained results can hardly be compared between centers mainly due to a lack of standardization of the lavage technique [3]. Instilled lavage volumes in the past ranged from 100 to 500 ml [1, 3, 4], Smaller vol umes are said to bear the risk that bronchial washings dominate the overall view [5, 6]. Larger volumes, al though are more representative of the alveolar spaces, may induce dangerous side effects, especially in highrisk patients [7, 8]. The aim of this study was to com-
Downloaded by: King's College London 137.73.144.138 - 11/27/2017 12:51:11 PM
Abstract. We have studied three different lavage procedures (100, 200 and 300 ml) in patients with pulmo nary sarcoidosis (stage I). The effect of bronchoalveolar lavage (BAL) on cell yield, lavage fluid recovery, dwelling time, lavage-induced arterial oxygen desaturation and occurrence of side effects was analyzed. The patients did not differ significantly in prelavage lung function and blood gas parameters. The lowest BAL re turn was seen in the 300-ml lavage procedure (49.5%), while the medium yielded over 70%. The lowest cell yield was seen in the BAL 100 g r o u p (10.4 x 106); the highest in the BAL 300 (19.4 x 106), but the latter did not differ significantly from BAL 200 (18.4 x 106). Dwelling time of the fluid differed only slightly between the small and middle volume lavage (average 3.2 vs. 3.9 min p < 0.01), but was significantly lower from the average dwelling time in the BAL 300 group (9.8 min, p < 0.001). Arterial oxygen desaturation was lowest in the BAL 100 and most pronounced in the large-volume lavage. Side effects were seen in all but I patient undergoing BAL 300. Cough was the most often reported side effect (9 patients); fever was observed in 6 patients, dyspnea in 4 (all undergoing large-volume lavage). Considering our results we do not think that it is justifiable to increase the volume of instilled fluid above 200 ml, because this may lead to serious side effects without increasing benefits. Using lower than 200 ml volumes decrease diagnostic yield although the risk of developing side effects is much lower.
Bronchoalveolar Lavage: Comparison of Three Commonly Used Procedures
pare the effect of the three most commonly used lavage techniques on bronchoalveolar lavage (BAL) return, arterial oxygen saturation, dwelling time and occurrence of side effects, as well as to propose a rea sonably safe lavage procedure, at the same time pro viding wide information.
Materials and Methods Sliuly Group 33 consecutive patients with stage I pulmonary sarcoidosis were included in the study. All were nonsmokers. They did not differ significantly in selected lung function and blood gas parameters (tables I, 2). They were allocated to three different lavage groups. In order to evaluate the role of fiberoptic bronchoscopy on the analysis of the effect o f BAL upon the studied parameters, a control group was made up of 13 consecutive patients undergoing routine diagnostic endoscopy due to suspected pulmonary malignancy.
73
Table 1. General characteristics of studied patients Group
Mean (±SD )age years'
Patients n
Sex ratio F:M
Control BAL 100 BAL 200 BAL 300
49.5 ± 4.5 32.4 ±2.4 41.1 ±2.4 41.0 ±4.4
13 8 18 7
1:12 3:4 1:1 1:3
1 All differences statistically significant (p < 0 .0 0 l), except between BAL 200 and 300 (NS).
Su.OV - SlcO?At m,n ~ s o f -------* 100
a s „o 2--------
where AS„.02 = desaturation rate in percent; S,cO°2 *• prelavage values o f arterial oxygen saturation: S,cO: BA1 m"' = lowest arterial
oxygen saturation reading during the lavage procedure. Side Effects. Each patient was asked to record all observed ad verse effects (i.e. cough, dyspnea, chills, myalgias, etc.) during the subsequent 24 h after the lavage procedure. Medical records were checked for the presence of fever (above 37.5°C)and other observa tions recorded by physicians during the 24 h after the lavage proce dure. Chest radiograms taken 24 h after BAL were reviewed for possible changes. Each patient's Slc0 2 was continuously monitored for 2 h after bronchoalveolar lavage. Additionally a recording of S(c0 2 24 h after BAL was carried out. These readings were related to prelavage values and expressed in percent according to the follow ing formula:
As„o-;jh =
S.cO? - S...O;4h
S„0:
-x 100
where AS,c0 224 h = desaturation rate 24 h following BAL or fiber optic bronchoscopy in percent: S,cO°2 = prelavage or prebronchoscopic values of S,cO:: S^.O:241' = Slt0 2 recorded 24 h after BAL or fiberoptic bronchoscopy. Statistical analysis was carried out using the t test.
Results There were no gross differences in lung function parameters and Pa02 in the analyzed patients that un derwent lavage. The BAL 300 group was the oldest of the lavaged patients, while the control group was the oldest of all (table 1). The poorest BAL return was noted in the large-volume lavage group (BAL 300), the highest in the medium lavage (BAL 200) - exceed ing 70% of the infused saline (differences statistically significant, p < 0.001; table 3).
Downloaded by: King's College London 137.73.144.138 - 11/27/2017 12:51:11 PM
Methods Premedication was not used in any patient. Only local anesthe sia was used (Xylocaine 2% - Astra). The maximal dose was ad justed individually but never exceeded 200 mg. Fiberoptic bron choscopy was performed using the transoral approach, with the pa tient in the supine position. Supplemental oxygen was not given in all evaluated patients. In all cases of diagnostic bronchoscopy for ceps biopsy was carried out. Olympus BF I TR endoscopes were used. Bronchoalveolar Lavage. Following spraying of the nasopha rynx, glottis, and larynx with 2% Xylocaine solution, the endo scope was inserted into the trachea. Immediately a 5-ml bolus of 2% Xylocaine was instilled to abolish cough reflexes. An additional 2 ml was always instilled into the orifice of the middle lobe bron chus. The tip of the endoscope was then advanced and wedged in a subsegmental bronchus (RB4 or RB5). in a position allowing visu alization of the distal airway. Two lavage procedures differed only in the volume of instilled saline. The BAL 100 (small volume) la vage consisted of introducing 5 portions of 20 ml each. The fluid was sucked back using the same syringe after each portion. The pressure was controlled to permit sufficient suction without collap sing the/bronchial walls. BAL 200 (middle volume) lavage was per formed similarly, 10 aliquots of 20 ml each were instilled. A differ ent technique was used for the BAL 300 (large volume) lavage. Saline was instilled in three 100-ml portions and suctioned back (immediately after each portion) into a sterile 500-ml side arm flask adapted to function as a suction trap. Electric suction was used, pressure being regulated by occlusion of the suction part of the en doscope, to prevent closure of the distal bronchial walls. The in stilled saline was always warmed up to the temperature of 37°C. BAL specimens were processed according to the method described by Hunninghake et al. [9]. BAL return was calculated in relation to the infused volume (in percent). Percutaneous Continuous Pulse Oximetry. An Ohmeda Biox 3700 continuous pulse oximeter was used. Continuous pulse oxime try was performed as described earlier by us [10]. Arterial oxygen desaturation was calculated from oximetric readings in relation to the prclavage values, according to the following formula:
74
Pirofcynski/Sliwihski/Radwan/Zielihski
Table 2. Selected lung function and blood gas parameters in studied patients Group
VC%PRED %
FEV,%PRED %
RV%TLC %
Raw cm H:0 / l/s
Pst/TLC cm H ,0
Cst/TGV
PaO, mm Hg
Control BAL 100 BAL 200 BAL 300
86.0 ±5.9 92.3 ± 5.6 97.3 ± 5.0 96.6 ±4.9
78.1 ±8.2 98.9 ±5.5 101.7 ±9.5 100.6 ±6.8
_
_
_
_
25.6 ± 3.1 24.2 ± 3.3 23.5 ±2.9
1.5 ±0.2 1.2 ±0.2 1.2 ±0.3
29.5 ±4.6 24.4 ±3.5 25.8 ±2.9
0.54 ±0.09 0.51 ±0.09 0.52 ± 0.05
78.7 ±2.5 84.8 ± 2.4 86.4 ±3.5 87.1 ±2.8
All differences between lavaged patients nonsignificant, differences between control and lavaged groups statistically significant ( p < 0.001).
Table 4. Side effects observed in control and lavage group Group
Group
BAL return' %
AS,cO; %
Cell yield' x I06
_
4.2 ±0.7 _ 6.3 ±0.9 10.3 ± 1.7 8.4 ± 1.2 18.4 ± 1.2 11.5 ± 1.7 19.4 ± 1.9
Dwelling time4, min
AS,c0 24h %
Side effects type
patients, n
Control
hemoptysis fever
3 2
0.01
BAL 100
cough
1
0.015
1 All differences statistically significant (p
![[Bronchoalveolar lavage].](/assets/img/hold.png)
