Chest Email Content Delivery
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sarodia, B. D.
Right arrow Articles by Arroliga, A. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sarodia, B. D.
Right arrow Articles by Arroliga, A. C.
(Chest. 2000;117:1043-1048.)
© 2000 American College of Chest Physicians

Does Pleural Fluid pH Change Significantly at Room Temperature During the First Hour Following Thoracentesis?*

Bipin D. Sarodia, MD; Lawrence S. Goldstein, MD; Daniel M. Laskowski, RPFT; Atul C. Mehta, MBBS, FCCP and Alejandro C. Arroliga, MD, FCCP

* From the Department of Pulmonary and Critical Care Medicine (Drs. Sarodia, Mehta, Arroliga, and Mr. Laskowski), Cleveland Clinic Foundation, Cleveland; and Northeastern Ohio Universities, College of Medicine, and Northside Medical Center (Dr. Goldstein), Youngstown, OH.

Correspondence to: Alejandro C. Arroliga, MD, FCCP, Cleveland Clinic Foundation, G-62, 9500 Euclid Ave, Cleveland, OH 44195; e-mail: arrolia{at}ccf.org


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Background: Usually, pleural fluid (PF) pH is measured immediately following thoracentesis, and if there is any delay in the measurement, the PF sample is preserved on ice.

Objective: To determine if PF pH changes significantly at room temperature during the first hour following thoracentesis.

Design: Prospective, self-controlled.

Setting: Tertiary care center.

Patients: All patients undergoing thoracentesis.

Measurements: The PF pH of a sample collected in an arterial blood gas syringe was measured immediately following thoracentesis by an arterial blood pH/gas analyzer. Additional measurements were made at 5, 15, 30, 45, and 60 min from the first pH measurement (pH0), maintained at room temperature.

Results: For 28 PF samples, pH0 (mean ± SD) was 7.351 ± 0.158, and the 60-min pH (pH60) was 7.359 ± 0.161. The mean difference between pH60 and pH0 was 0.008 ± 0.026, which was not significant, either clinically or statistically (p = 0.13). Similarly, the interim pH values (for measures at 5, 15, 30, 45 min after pH0) were not significantly different from pH0 (mean differences, 0.002, 0.003, 0.005, and 0.004, respectively; p values, 0.51, 0.21, 0.06, and 0.22, respectively).

Conclusions: The pleural fluid pH of a sample preserved at room temperature does not change significantly during the first hour following thoracentesis. Hence, contrary to the common medical practice, there is no need to perform the pH measurement within minutes after thoracentesis and to preserve a pleural fluid sample on ice.

Key Words: empyema • exudate • hydrogen ion concentration • pleural effusion • pleural fluid • specimen handling • temperature • time factor • transudate


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Multiple tests are performed on pleural fluid (PF) samples obtained by diagnostic thoracentesis to help manage patients with pleural effusions. The protein and lactate dehydrogenase (LDH) concen- trations of PF are measured to classify a pleural effusion as exudative or transudative. PF pH is measured to detect PF acidosis (pH < 7.30). These measurements have important diagnostic, prognostic, and therapeutic implications.1 2 3 4 5 6 7 8 9

According to common medical practice and recommendations for PF pH measurement, the PF is collected anaerobically in a heparinized syringe and the PF pH measured immediately following thoracentesis. When delay is unavoidable, the sample is preserved on ice to prevent spontaneous acid generation and a false pH result.6 8 9 10 11 There is no strong scientific evidence in published English-language literature to substantiate these recommendations. We have shown earlier that even if the fluid is collected in a regular 30- or 60-mL syringe and then transferred to a heparinized syringe, the PF pH does not change significantly.12

The primary aim of this study was to determine if the change in PF pH at room temperature during the first hour following thoracentesis is statistically or clinically significant. The secondary aim was to determine how soon after thoracentesis any important change occurs. Our hypothesis was that the change would be statistically insignificant, or if statistically significant, then clinically insignificant. We selected 1 h as the maximum time, because common medical practice presumed that PF pH would change within 1 h. Also, we believed that eliminating the rush to measure PF pH within few minutes after thoracentesis might make the procedure more convenient.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Conclusion
 References
 
All patients, either outpatients or inpatients from any unit or floor of our institution, a tertiary care academic medical center, who were undergoing thoracentesis for any indication were eligible for this study. The patients were enrolled between May 1997 and February 1999 whenever one of the authors (BDS, LSG, or DML) was available to measure PF pH. The fluid was also analyzed by other laboratory tests, at the discretion of the ordering physician.

Measurement of PF pH
A 3-mL PF sample was collected in a heparinized syringe from the 30- or 60-mL syringe used for thoracentesis. Air bubbles were avoided by transferring the fluid slowly, and if present, they were removed by gently tapping the syringe. The syringe was kept at room temperature and tightly closed to avoid exposure to air throughout the study period, except during sampling for pH measurement. The sample was transported to the laboratory, and the first PF pH measurement (pH0) was measured within 15 min after thoracentesis. Additional pH measurements were made at 5 min after the first measurement (pH5), 15 min after the first measurement (pH15), 30 min after the first measurement (pH30), 45 min after the first measurement (pH45), and 60 min after the first measurement (pH60). All these PF pH measurements were done once at each given time schedule.

The blood pH/gas analyzer ABL300 acid-base laboratory (Radiometer; Copenhagen, Denmark) was used for PF pH measurement. The instructions of the manufacturer were strictly followed.13 Just before each PF pH measurement, a 1-point calibration and a test for pH-electrode sensitivity (92 to 104%) were performed, and the sample was thoroughly mixed. Heat exchangers in the analyzer machine were used to bring the sample to 37°C before pH measurement, because PF pH changes inversely with its temperature. As per the specifications of the manufacturer, the inaccuracy (mean difference from the estimated pH by a reference method) of this machine is - 0.003 to + 0.006 for pH from 6.95 to 7.70. The repeatability (standard deviation of repeated measurements) is 0.002. Overall imprecision for a single pH measurement (range of standard deviations due to day-to-day variation, and variations in the instrument, standard solutions, and reference methods) is 0.005 to 0.007.13

Exclusion Criteria
A PF pH value was excluded if it was not measured within 15 min following thoracentesis for pH0, and within 5 min of the scheduled time for other measurements. A PF sample was excluded if the pH data for that sample did not include pH0, pH60, and at least two other measurements, because of any reason including insufficient amount of the sample, unavailability of pH analyzer machine or the personnel for PF pH measurement, error message in the machine, or machine malfunction due to sample clotting.

Statistical Methods
The differences between pH0 and other PF pH values were evaluated for a statistical and clinical significance. A p value < 0.05 for a two-tailed paired t test was considered statistically significant. Pearson correlation coefficients were calculated for comparison of the mean pH0 and other PF pH values. Microsoft Excel 97 (Microsoft Corporation; Redmond, WA) and Complete Statistical System (Statsoft; Tulsa, OK) were used for the data analysis. A pH difference > 0.05 was considered clinically significant. The statistical power to detect this difference was estimated using the UnifyPow macro for SAS version 6.12 (SAS Institute; Cary, NC) based on the standard deviations and the Pearson product-moment correlation between the pH values.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Conclusion
 References
 
Thirty-six PF samples were obtained. Eight samples were excluded for the following reasons: (1) pH0 was not measured within 15 min following thoracentesis because of delay in transportation of the sample to the laboratory (n = 4); (2) pH60 was not measured because of insufficient amount of the sample (n = 1); and (3) more than two other PF pH values (other than pH0 and pH60) for a given sample were not measured within 5 min of the scheduled time because of the unavailability of pH analyzer machine (n = 1) or because of the machine malfunction due to sample clotting (n = 2).

The patient and pleural effusion characteristics for the remaining 28 PF samples are shown in Table 1 . The etiologies of the pleural effusions in these patients were as follows: primary thoracic malignancy (n = 11), with adenocarcinoma (n = 4) and other non-small cell carcinoma of lung (n = 1), esophageal carcinoma (n = 3), mesothelioma (n = 1), Hodgkin’s lymphoma (n = 1), and non-Hodgkin’s lymphoma (n = 1); metastatic extrathoracic malignancy (n = 5), with breast adenocarcinoma (n = 2), melanoma (n = 1), ovarian carcinoma (n = 1), and right hypernephroma (n = 1); parapneumonic effusions (n = 4); congestive heart failure (n = 3); ARDS (n = 2); traumatic hemothorax (n = 1); chylothorax (n = 1); and unknown (n = 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Patient and Pleural Effusion Characteristics (n = 28)*

 
The pH0 (mean ± SD) was 7.351 ± 0.158, and pH60 was 7.359 ± 0.161 (Tables 2 , 3 ). The change in PF pH over 1 h was neither clinically nor statistically significant (mean difference ± SD, 0.008 ± 0.026; p = 0.13). There was a strong positive correlation between pH0 and pH60 (correlation coefficient, 0.987; p < 0.001; Fig 1 ). The statistical power to detect a difference of 0.05 between pH0 and pH60 (based on the difference in their mean ± SD and correlation coefficient) was > 95%. Similarly, for the subgroup with PF acidosis (8 of 28 samples), the change in PF pH over 1 h was not significant (NS) (pH0, 7.176 ± 0.209 vs pH60, 7.178 ± 0.203; mean difference ± SD, 0.002 ± 0.019; p = 0.83). Even a very low PF pH (pH0, 6.664) in one patient (patient 16 in Table 2 ) did not change following 1 h at room temperature (pH60, 6.680; change, 0.016; Fig 1 ).


View this table:
[in this window]
[in a new window]

  		Table 2. PF pH, Cell Count, and Chemistry Results for Individual Patients (n = 28)*

 


View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Scatter diagram with a trend line of linear regression showing strong correlation (r = 0.987; p < 0.001) between PF pH values measured (1) immediately following thoracentesis (pH0), and (2) after 60 min of the sample preservation at room temperature (pH60). The differences between pH0 and pH60 were not clinically or statistically significant, for either the whole group (n = 28) or for the fluids with acidosis (pH0 < 7.30; n = 8).

 
The changes in pH5, pH15, pH30, and pH45 from pH0 were also NS (mean differences, 0.002, 0.003, 0.005, and 0.004, respectively; p values, 0.51, 0.21, 0.06, and 0.22, respectively). For 28 PF pH values of pH0 and pH60, the sample sizes for pH5, pH15, pH30, and pH45 were 24, 24, 24, and 25, respectively. Hence, the comparison of each of these interim pH values and pH0 was done after excluding the pairs with missing data. There was a strong positive correlation of pH5, pH15, pH30, and pH45 with pH0 (correlation coefficients 0.997, 0.998, 0.997, and 0.995, respectively; p < 0.001 for each). The changes in pH5, pH15, pH30, pH45, and pH60 from pH0 were clinically insignificant (<= 0.05) in 97.6% (122 of 125) of values compared. Only 2.4% (3 of 125) of changes (-0.066, 0.060, and 0.066; all for pH60) were > 0.05 and therefore clinically significant.

Table 2 shows the individual values of pH0 and pH60, the difference between pH0 and pH60, the cell counts and chemistry of PF, and the correlation of pH0 to these variables for all 28 patients. The pH0 value correlated inversely with neutrophil count (expressed as percent of total WBC count; r = -0.41; p = 0.03) and LDH concentration in the PF (r = -0.63; p < 0.001), but did not correlate with the total WBC count, lymphocyte count, protein, or glucose concentration in the PF (p > 0.05).


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Conclusion
 References
 
In the present study, we have demonstrated that there is no clinically or statistically significant change in PF pH during the first hour following thoracentesis in various types of pleural effusions, even if the PF is maintained at room temperature (Fig 1 and Tables 2 , 3 ). The pH60 was expected to be lower than pH0 due to acid generation in the fluid, but instead the mean value was insignificantly higher by 0.008 pH units (Tables 2 , 3) . The mean difference was just above the upper level of imprecision (0.007) of the pH measurement by the analyzer machine. The changes in interim PF pH values (pH5, pH15, pH30, and pH45) from pH0 were also NS. The changes from pH0 for 97.6% of PF pH values (other than pH0) were clinically insignificant.


View this table:
[in this window]
[in a new window]

  		Table 3. Comparison of PF pH at 0 and 60 Min (n = 28)

 
One might hypothesize that the PFs with a low pH would have a tendency to generate more acid over 60 min. However, this study suggests otherwise. The pH of the eight PF samples with low pH (pH < 7.30) also did not change significantly (mean difference, 0.002 ± 0.019; p = 0.83) at room temperature during the first hour following thoracentesis. Even a very low baseline PF pH did not change significantly (pH0, 6.664; patient 16; Fig 1 ; Table 2 ). The patient with this PF pH was a 61-year-old man who had squamous cell carcinoma of right middle lobe and associated right pleural effusion with very high PF LDH (1,941 U/L), protein (6.1 g/dL), and neutrophil count (63%), and negative microbiologic cultures and cytology. His pleural fluid was considered to be parapneumonic. His malignancy was staged as III B, and he underwent right pneumonectomy without immediate complications.

There are important clinical implications of PF pH measured accurately. A single PF pH reflects a steady state of multiple metabolic and kinetic factors. The proposed mechanisms of PF acidosis include multiple factors other than acid generation by the PF per se.6 The PF pH is directly related to the arterial blood pH, and in transudative effusions, it is usually higher than the simultaneous blood pH.2 The common clinical conditions causing an exudative pleural effusion with a PF pH of < 7.20 include empyema, complicated parapneumonic effusion, malignant pleural disease, tuberculous pleuritis, esophageal rupture, hemothorax, and collagen vascular diseases, such as rheumatoid and lupus pleuritis.3 4 8 9 Parapneumonic effusions with PF acidosis usually do not resolve spontaneously, may loculate, and require a chest tube thoracostomy, despite not being strictly classified as empyemas.2 4 PF acidosis in malignant pleural effusions is associated with higher positive rates in PF cytology results, shorter survival, and poor response to pleurodesis.5

Changes in PF pH with time and preservation of the sample at different temperatures has been evaluated in three studies. In the first study of 23 samples preserved at 37°C incubation, the mean reduction in PF pH was 0.54 (range, 0 to 1.8) after 24 h.6 None of the samples showed an increase in PF pH. The degree of acid generation was not related to the baseline pH, but it was higher in PFs with higher leukocyte count and lower oxygen tension. In addition, it was greatest in empyema, intermediate in carcinomatous effusions, and minimal in rheumatoid pleuritis. Although PF pH was also measured at 3, 6, and 12 h from the time of incubation, these results were not reported.6 In a second study, the pH of eight PF samples maintained anaerobically at 0°C for >= 8 h did not change > 0.02.2 In a third study of 25 exudative, nonpurulent pleural effusions refrigerated at 5°C, PF pH did not change significantly even after 24 h (mean increase, 0.035; p = 0.17).11 However, to our knowledge, there is no report in English-language literature describing the occurrence or magnitude of change in PF pH at room temperature during the first hour following thoracentesis.

Reports comparing the accuracy of the blood gas machine, pH meter, and pH indicator strips have demonstrated that only the blood gas machine is sufficiently accurate.10 11 14 Because we could find no change over time even using the blood gas machine, the most accurate method, our results indicate a genuine absence of change, not a measurement error.

One of the minor limitations of this study is that the number of patients studied is small. However, the statistical power of detecting the clinically significant difference of 0.05 is > 95%. A larger sample size might have been helpful in subgroup analysis to further support our conclusions. Another limitation is that there may be a small selection bias due to exclusion of two grossly purulent samples that clogged the fluid in the analyzer machine. However, some authors believe that determining PF pH may not change the management of grossly purulent effusions, because they all require drainage by chest tube thoracostomy.11

Although our study was not designed to assess the correlation between PF pH and other characteristics of PF, it was found that pH0 correlated (p < 0.05) inversely with neutrophil count (expressed as percent of total WBC count) and LDH concentration in PF, but had no significant correlation with total WBC or other cell counts and other chemistry values (p > 0.05; Table 2 ). One study investigating the relation between the baseline PF pH and chemistry found similar inverse correlation for LDH concentration, and also for the ratio of PF glucose to serum glucose concentration.7


   Conclusion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
References
 
The PF pH of a sample preserved anaerobically (exposed to air only during transfer from the syringe to a blood pH/gas analyzer) at room temperature does not change significantly during the first hour following thoracentesis. Hence, contrary to the common medical practice, the sample need not be rushed to the laboratory within few minutes after thoracentesis, but could be taken to the laboratory at leisure to have the pH measured any time during the first hour without being preserved on ice. This is true for a variety of pleural effusions, both transudative and exudative, and with either normal or low pH. These changes may make the procedure of PF pH measurement more convenient without compromising the quality of the pH results.


   Acknowledgements
 
The authors thank David R. Nelson, MS, Senior Biostatistician, Department of Biostatistics and Epidemiology, The Cleveland Clinic Foundation, for his valuable statistical guidance; and Jessica Ancker, BA, Medical Editing Services, The Cleveland Clinic Foundation, for her immense assistance in editing the manuscript.


   Footnotes
 
Abbreviations: LDH = lactate dehydrogenase; NS = not significant; PF = pleural fluid; pH0 = the first PF pH measurement; pH5 = pH measurement made 5 min after the first measurement; pH15 = pH measurement made 15 min after the first measurement; pH30 = pH measurement made 30 min after the first measurement; pH45 = pH measurement made 45 min after the first measurement; pH60 = pH measurement made 60 min after the first measurement

Received for publication August 25, 1999. Accepted for publication December 21, 1999.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Light, RW, MacGregor, MI, Luchsinger, PC, et al (1972) Pleural effusions: the diagnostic separation of transudates and exudates. Ann Intern Med 77,507-513
  2. Light, RW, MacGregor, MI, Ball, WC, Jr, et al (1973) Diagnostic significance of pleural fluid pH and PCO2. Chest 64,591-596[Abstract/Free Full Text]
  3. Good, JT, Jr, Taryle, DA, Maulitz, RM, et al (1980) The diagnostic value of pleural fluid pH. Chest 78,55-59[Abstract/Free Full Text]
  4. Potts, DE, Levin, DC, Sahn, SA (1976) Pleural fluid pH in parapneumonic effusions. Chest 70,328-331
  5. Sahn, SA, Good, JT, Jr (1988) Pleural fluid pH in malignant effusions: diagnostic, prognostic and therapeutic implications. Ann Intern Med 108,345-349
  6. Taryle, DA, Good, JT, Jr, Sahn, SA (1979) Acid generation by pleural fluid: possible role in the determination of pleural fluid pH. J Lab Clin Med 93,1041-1046[Medline]
  7. Potts, DE, Willcox, MA, Good, JT, Jr, et al (1978) The acidosis of low-glucose pleural effusions. Am Rev Respir Dis 117,665-671[ISI][Medline]
  8. Houston, MC (1987) Pleural fluid pH: diagnostic, therapeutic, and prognostic value. Am J Surg 154,333-337[CrossRef][Medline]
  9. Light, RW (1995) Pleural diseases 3rd ed. ,50-52 Williams & Wilkins Baltimore, MD.
  10. Cheng, D, Rodriguez, M, Rogers, J, et al (1998) Comparison of pleural fluid pH values obtained using blood gas machine, pH meter, and pH indicator strip. Chest 114,1368-1372[Abstract/Free Full Text]
  11. Chandler, TM, McCoskey, EH, Byrd, RP, Jr, et al (1999) Comparison of the use and accuracy of methods for determining pleural fluid pH. South Med J 92,214-217[CrossRef][ISI][Medline]
  12. Goldstein, LS, McCarthy, K, Mehta, AC, et al (1997) Is direct collection of pleural fluid into a heparinized syringe important for determination of pleural pH? A brief report. Chest 112,707-708[Abstract/Free Full Text]
  13. User’s handbook: pH/blood gas analyzer ABL 300 acid-base laboratory. Copenhagen, Denmark; Radiometer, 1984
  14. Lesho, E, Roth, BJ (1997) Is pH paper an acceptable, low-cost alternative to the blood gas analyzer for determining pleural fluid pH? Chest 112,1291-1292[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sarodia, B. D.
Right arrow Articles by Arroliga, A. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sarodia, B. D.
Right arrow Articles by Arroliga, A. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS