HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Mathew Ninan K. Eric Sommers Rodney J. Landreneau James D. Luketich Peter F. Ferson Robert J. Keenan Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ninan, M. Articles by Keenan, R. J. Search for Related Content PubMed PubMed Citation Articles by Ninan, M. Articles by Keenan, R. J. Related Collections Related Article

Ann Thorac Surg 1997;64:328-332
© 1997 The Society of Thoracic Surgeons

---

Original Articles: General Thoracic

Standardized Exercise Oximetry Predicts Postpneumonectomy Outcome

Section of Thoracic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania

	Abstract

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Background. We have developed a safe, simple, and easily performed standardized exercise oximetry outpatient test to assess patients undergoing lung resections. We studied its ability to predict outcome after pneumonectomy in 46 consecutive patients over a 5-year period.

Methods. Room air oximetry is initially performed at rest. The patient then begins to exercise on a stair-stepper apparatus (Stamina Stepper), which provides uniform resistance to stepping. Oxygen saturation values are noted at 10, 20, and 30 steps, equivalent to climbing three flights of stairs. Group 1 consisted of the patients who either had a resting saturation less than 90%, or desaturation greater than or equal to 4% during exercise. Group 2 consisted of all patients who had a preoperative forced expiratory volume in 1 second of 60% or less. Group 3 consisted of all patients who had a predicted postoperative forced expiratory volume in 1 second of 40% or less. Group 4 consisted of patients who had a predicted postoperative diffusing capacity of 40% or less.

Results. There were four deaths (8.6%), 12 patients (26%) remained in the intensive care unit 4 or more days, and 11 patients (23%) suffered major morbidity. Desaturation during exercise (group 1) significantly predicted longer intensive care unit stay (p = 0.0002) and incidence of major morbidity (p < 0.0001). Groups 2, 3, and 4 were not significantly predictive of either longer intensive care unit stay or major morbidity.

Conclusions. Standardized exercise oximetry performed in the outpatient facility is highly predictive of major morbidity and prolonged intensive care unit stay after pneumonectomy.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
See also page 332.

The pulmonary function test most commonly used to predict outcome after lung resections has been spirometry. However, many studies have documented the poor predictive value of this modality [1–4]. Hence a range of other tests have been evaluated to identify the patient with a high risk for developing complications after lung resection. These have ranged from simple walking exercise studies [3], exercise oximetry [4, 5], and determination of carbon monoxide lung diffusing capacity [6, 7], to more sophisticated and more expensive maneuvers such as maximum oxygen uptake [8] and pulmonary vascular resistances [9]. Although maximal oxygen uptake and diffusing capacity estimations have been shown to be predictive of morbidity after lung resections, an ideal screening test that predicts outcome accurately after major pulmonary resections like pneumonectomy is lacking. The ideal screening test should provide a sensitive, specific, and cost-effective identification of the high risk patient, and this subgroup may require further sophisticated tests. Although exercise oximetry has been shown to be useful in predicting outcome, a standardized version of the test is not available. We report our use of a simple and useful outpatient screening test using exercise oximetry as a primary step in determining the risk for postoperative morbidity after pneumonectomy.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Patient Demographics
The in-hospital and outpatient records of 47 consecutive patients who presented to our outpatient facility from January 1991 to January 1996 and eventually underwent pneumonectomy were reviewed. All patients had a diagnosis of non-small cell lung cancer, either known at the time of testing, or confirmed before the operation. All these patients underwent a standardized exercise oximetry test and spirometry in the outpatient facility. One patient could not perform the oximetry test owing to an old knee injury, leaving 46 patients who formed the study group. The study group had ages ranging from 38 to 82 years (mean 60.1 years). There were 13 women and 33 men. There were 22 right-sided and 24 left-sided pneumonectomies. Eight patients were in stage I, 20 patients in stage II, and 18 patients in stage IIIa non-small cell lung cancer. Six of the eighteen patients in stage IIIa had preoperative chemotherapy after outpatient assessment.

Exercise Oximetry
Exercise oximetry was performed using a stair-stepper apparatus (Stamina Stepper; Stamina Products Inc, Springfield, MO) by the same respiratory technician. The Stamina Stepper is an exercise apparatus (commonly used in exercise gymnasiums) with a uniform stepping height and uniform resistance to stepping. It mimics stair-climbing while exercising. Initially, resting arterial oxygen saturation was measured using a finger oximeter (Nellcor N-100 pulse oximeter; Nellcor Inc, Hayward, CA). Then the patient began to exercise on the stair stepper and finger oxygen saturation was noted after completing 10, 20, and 30 steps on the apparatus. This is roughly equivalent to climbing three flights of stairs. The patient did not have a rest period between the steps. The step rate was standardized by requesting the patient to time the steps to a visible pendulum. Further, no allowance was made in the step rate for the age and sex of the patient. All 46 patients completed the full test.

Spirometry
All patients underwent standard spirometry using a spirometer with analog output to a computer. All spirometric values were the best of three attempts and were also expressed as a percentage based on the age, sex, height, and weight of the patient. Suboptimal efforts on the spirometer were discarded. Predicted postoperative forced expiratory volume in the first second (%FEV₁-ppo) was calculated from the quantitative ventilation perfusion scans or by calculation depending on which lung was removed. FEV₁-ppo was calculated by the following formula: FEV₁-ppo = preoperative FEV₁ x (1 - fractional perfusion of lung to be resected).

Other Pulmonary Function Tests
All patients who had radiologic findings suggesting the need for a pneumonectomy underwent lung diffusing capacity for carbon monoxide (D_Lco) estimations and quantitative ventilation-perfusion lung scintiscanning during their outpatient visit. Lung diffusing capacity measurements were performed by the single breath method during a 10-second breath-holding maneuver. This value was then corrected for hematocrit and lung volumes and also expressed as a percentage for that individual. Predicted postoperative values were calculated by the split function of the lungs as determined by the quantitative ventilation-perfusion scan.

Post Operative Morbidity
Six objective outcomes were taken to indicate major morbidity in the postoperative period. These were pneumonia, prolonged ventilation of more than 12 hours after intensive care unit (ICU) admission, adult respiratory distress syndrome in the remaining lung, reintubation, requirement of home oxygen, and death. All the major morbidities were examined together as a composite end point. As evident, all the morbidities examined are major or life-threatening outcomes after pneumonectomy. An ICU stay of 4 or more days was examined as an independent variable of poor outcome.

Surgical Management
All 46 patients underwent standard right- or left-sided pneumonectomy by posterolateral thoracotomy. Twenty patients had a muscle-sparing thoracotomy, and in the others the latissimus dorsi was divided and the serratus anterior muscle was either spared or divided. There were no reoperations in the immediate postoperative period. One patient underwent venovenous extracorporeal membrane oxygenation for postpneumonectomy pulmonary edema. He was weaned from extracorporeal membrane oxygenation in 26 days. Twelve patients had a thoracic epidural catheter placed for postoperative analgesia, and the others had intravenous analgesia only.

In our institution, patients undergoing pneumonectomy are managed for the first 12 to 24 hours in the ICU with a 1:2 nurse to patient ratio. Patients are then transferred to the ward floor, where they remain till discharge. There is no intermediate care unit in our hospital. Patients are transferred to the ward floor from the ICU after discussion between the surgical and critical care teams. Patients are usually discharged on the seventh postoperative day.

High-risk Subgroups
Patients were divided into four high-risk subgroups. Group 1 were all those who desaturated at least 4% from resting values at any time during or immediately after the oximetry test. Also included in this category were those who began the test with a resting saturation of less than 90% (11 patients). Group 2 consisted of all patients who had a preoperative FEV₁ less than 60% (6 patients). Group 3 were those patients who had a predicted postoperative FEV₁ (%FEV₁-ppo) less than 40% (21 patients). Group 4 consisted of all patients with a predicted postoperative D_Lco (%D_Lco-ppo) less than 40% (15 patients). As evident, there was overlap between the patient groups, with 6 patients who desaturated having a %FEV₁-ppo of less than 40%.

Statistical Analysis
Analysis was done using JMP software (SAS Inc., Cary, NC) on a Power PC Macintosh. All variables were converted to categorical data and analyzed using the ² test. Significance was assessed using the Fisher's two-tailed exact test. Sensitivity and specificity for these groups were calculated by the same program.

	Results

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Morbidity and Mortality
There were four deaths within 30 days of the pneumonectomy (30 day mortality = 8.6%). Three of these deaths were caused by pneumonia and respiratory failure and the other was the result of severe postoperative adult respiratory distress syndrome. Eleven patients had major morbidity postoperatively (23%). There were six instances of pneumonia, five instances of prolonged ventilation, 2 instances of adult respiratory distress syndrome, two reintubations, and four patients needed home oxygen. Twelve patients (26%) remained in the ICU 4 or more days.

Pulmonary Function Data
Ten patients desaturated 4% or more during or immediately after the stair-stepper test, and 1 patient entered the test with a resting saturation of 88% and desaturated 2% during the test (group 1), 5 patients desaturated 4%, 1 patient desaturated 5%, 2 desaturated 6%, 1 desaturated 7%, and 1 desaturated 14%. In contrast, 5 patients had a higher saturation after exercise than at rest and 10 patients had the same saturation before and after the test. The other 10 patients desaturated between 1 and 3% during the test.

Preoperative FEV₁ percentage ranged from 38% to 122% (mean 80.9%). Six patients had an FEV₁% less than 60% preoperatively (group 2). Predicted postoperative FEV₁ (%FEV₁-ppo) ranged from 20% to 68% (mean 40.3%). Twenty-one patients had a %FEV₁-ppo less than 40% (group 3). The predicted postoperative D_Lco (%D_Lco-ppo) ranged from 23% to 59% (mean 43.4%). Fifteen patients had a %D_Lco-ppo of less than 40%.

Incidence of Major Morbidity and ICU Stay with Exercise Desaturation (Group 1)
Group 1 patients were significantly predictive of incidence of the six major morbidities (p < 0.0001, Fisher's two-tail exact test). The exercise oximetry test was 90.91% sensitive and 77.14% specific for major morbidity after pneumonectomy. Seven of the eleven patients in the high-risk group 1 remained 4 or more days in the ICU. When assessed by the Fisher's two-tail exact test, group 1 was highly predictive for prolonged ICU stay (p = 0.0002). The sensitivity of the exercise oximetric test for prolonged ICU stay was 72% and the specificity was 88.5%. Figures 1 and 2 show the differences in postoperative morbidity and ICU stay of patients who desaturated more or less than 4% during the exercise oximetric test.

View larger version (13K): [in this window] [in a new window]   Fig 1. . Incidence of combined major morbidity in patients desaturating less than or more than 4% during exercise oximetry.

View larger version (14K): [in this window] [in a new window]   Fig 2. . Comparison of mean intensive care unit (ICU) stay in patients in high and low risk groups of exercise desaturation.

Similarly, a low preoperative FEV₁% (group 2) was not predictive of the incidence of major morbidity in the postoperative period (p = 0.6656, Fisher's exact test) with a sensitivity of 50% and a specificity of 62.5%. This was also true if a low predicted postoperative value of less than 40% was used (group 3). This group was not predictive of major morbidity (p = 0.7693, Fisher's exact test) with a sensitivity for this test of only 36% and a specificity of 58%. Figures 3 and 4 show the difference in postoperative outcome of patients with an FEV₁-ppo of more or less than 40%.

View larger version (19K): [in this window] [in a new window]   Fig 3. . Mean intensive care unit (ICU) stay of patients with a predicted postoperative forced expiratory volume in 1 second (%FEV1-ppo) of more or less than 40%.

View larger version (19K): [in this window] [in a new window]   Fig 4. . Incidence of combined major morbidity in patients with a predicted postoperative forced expiratory volume in 1 second (%FEV1-ppo) of more or less than 40%.

	Comment

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
At present, criteria used to select patients for major pulmonary resections are the clinical assessment, spirometric values, and in selected cases, further pulmonary and cardiac assessment. However, traditional criteria used to predict risk after lung resections provide only a modest ability to do so. This has stimulated a search for better preoperative assessment of these often marginal patients. A large amount of literature has accumulated concerning patient selection and the best means to do so. Preoperative factors that have been shown to be significant are preoperative and postoperative FEV₁ values, maximum ventilatory volume, arterial oxygen saturation during exercise, maximum oxygen consumption, mean pulmonary artery pressures, pulmonary arterial carbon dioxide levels, D_Lco, and pulmonary vascular resistance. Studies of patients undergoing both lobectomy and pneumonectomy have found spirometry to be of no benefit in predicting outcome after these major resections [1, 2]. However, Markos and colleagues [5] prospectively assessed 53 patients undergoing lung resections and found predicted postoperative FEV₁ expressed as a percentage (%FEV₁-ppo), postoperative predicted D_Lco expressed as a percentage (%D_Lco-ppo), and exercise-induced arterial oxygen desaturation to be the best predictors of postoperative outcome for the group as a whole, with %FEV₁-ppo being the single best predictor after pneumonectomy. This group consisted of only 18 pneumonectomies. Ferguson and associates [6] retrospectively reviewed 237 patients (including 73 pneumonectomies) and found that the single-breath diffusing capacity for carbon monoxide was the sole predictor of major morbidity and mortality. This study did not examine arterial oxygen saturation as a variable. Kearney and coworkers [10] prospectively studied 331 patients undergoing various types of wedge resections, lobectomy, and pneumonectomy (n = 64) and found that a low predicted postoperative FEV₁ (%FEV₁-ppo) was the best predictor of postoperative outcome. A form of nonstandardized exercise oximetry was also examined in this study, and the authors found that desaturation with exercise did not predict outcome. However, supraventricular tachycardia was the most common morbidity reported in this study, an outcome that is unlikely to be predicted by exercise desaturation.

In all, four studies have included exercise oximetry in the assessment of risk after all lung resections. Two of them [5, 10] have been discussed. Holden and colleagues [3] reported on a small number of patients undergoing a 6-minute walk with finger pulse oximetry; those with a 6-minute walk distance of greater than 100 feet without desaturation were predictive of successful surgical outcome. A recent study [4] compared exercise oximetry to spirometry retrospectively in 299 patients undergoing wedge resections, lobectomies, and pneumonectomies. The standardized exercise oximetry used in this study was a 150-meter "walk around the ward" and then climbing two flights of stairs. They found that exercise oximetry predicted postoperative outcome better than spirometry, though the sensitivity and specificity of both these modalities were low. As evident from the previous discussion, most studies have combined all varieties of lung resections for analysis. We were of the opinion that a test of exercise desaturation would be more predictive of outcome after a major resection such as a pneumonectomy, rather than after a wedge resection, and that the test would have to be performed in a more standardized fashion than previously. Interestingly, the data from the study of Markos and associates [5] suggested that in patients with a low maximum oxygen consumption (<15 mL · kg^-1 · min^-1), 63% also had exercise arterial oxygen desaturation.

All these factors led us to initiate a standardized form of exercise oximetry testing using the stair-stepper apparatus since 1991. The object of our study was to assess the immediate postoperative outcome of patients who desaturated with exercise, as compared to patients who had low predicted spirometric values and low diffusing capacities. The postoperative outcomes examined in our study were all life-threatening outcomes after pneumonectomy, mortality also being included. Thus, relatively minor morbidity such as new bronchodilator use, wound infection, and supraventricular arrhythmia were not used. A prolonged ICU stay was examined independently. Home oxygen use after discharge from hospital was used as it suggested significant pulmonary compromise after pneumonectomy. The spirometric analysis consisted of the two most predictive factors reported after lung resections, ie, a low predicted FEV₁ expressed as a percentage, and a low predicted postoperative FEV₁ expressed as a percentage [3, 5, 10]. Overall, we found that desaturation during exercise significantly predicted prolonged ICU stay and major morbidity. The FEV₁% and %FEV₁-ppo did not predict either of these postoperative outcomes. We were also unable to show a low predicted postoperative D_Lco percentage (< 40%) to be predictive of either prolonged ICU stay or major morbidity. There is no clear evidence in the literature that exercise desaturation correlates with a low diffusing capacity. However, previous studies have shown a clear relationship between a low diffusing capacity and poor postoperative outcome after lung resections [6, 7].

Our study did not examine the potentially important factor of rate of recovery from exercise desaturation. This could reveal more about the respiratory reserve of the patient. Also, the patient numbers were too small to permit a meaningful correlation between the degree of exercise and desaturation and postoperative outcome. The numbers of individual morbidities were insufficient to permit an assessment of risk for each morbidity from desaturation. Also, the majority of patients did not have maximum oxygen consumption assessed, thus comparison between desaturation and this maximum oxygen consumption was not possible. The situation could be clarified by a prospective randomized trial using our form of standardized exercise oximetry, and this study could also compare exercise oximetry to both maximum oxygen consumption and diffusing capacity. In the interim we continue to believe that standardized stair-stepper exercise oximetry remains the most useful tool to screen patients for pneumonectomy, being more sensitive and specific than spirometry. In addition, it is a cost-effective test and can be performed in a standardized manner in the outpatient facility.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Presented at the Forty-third Annual Meeting of the Southern Thoracic Surgical Association, Cancun, Mexico, Nov 7–9, 1996.

Address reprint requests to Dr Ninan, Suite C-700, Presbyterian University Hospital, 200 Lothrop St, Pittsburgh, PA 15213.

	References

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 

1. Keagy BA, Schorlemmer GR, Murray GF, Starek PJK, Wilcox BR. Correlation of preoperative pulmonary function testing with clinical course in patients after pneumonectomy. Ann Thorac Surg 1983;36:253–7.[Abstract]
2. Didolkar MS, Moore RH, Takita H. Evaluation of the risk in pulmonary resection for bronchogenic carcinoma. Am J Surg 1974;127:700–3.[Medline]
3. Holden DA, Rice TW, Stelmach K, Meeker DP. Exercise testing, 6-min walk, and stair climb in the evaluation of patients at high risk for pulmonary resection. Chest 1992;102:1774–9.[Abstract/Free Full Text]
4. Rao V, Todd TRJ, Kuus A, Buth KJ, Pearson FG. Exercise oximetry versus spirometry in the assessment of risk prior to lung resection. Ann Thorac Surg 1995;60:603–9.[Abstract/Free Full Text]
5. Markos J, Mullan BP, Hillman DR, et al. Preoperative assessment as a predictor of mortality and morbidity after lung resection. Am Rev Respir Dis 1989;139:902–10.[Medline]
6. Ferguson MK, Little L, Rizzo L, et al. Diffusing capacity predicts morbidity and mortality after pulmonary resection. J Thorac Cardiovasc Surg 1988;96:894–900.[Abstract]
7. Ferguson MK, Reeder LB, Mick R. Optimizing selection of patients for major lung resection. J Thorac Cardiovasc Surg 1995;109:275–83.[Abstract/Free Full Text]
8. Walsh GL, Morice RC, Putnam JB Jr., et al. Resection of lung cancer is justified in high-risk patients selected by exercise oxygen consumption. Ann Thorac Surg 1994;58:704–11.[Abstract]
9. Fee HJ, Holmes EC, Gewirtz HS, Ramming KP, Alexander JM. Role of pulmonary vascular resistance measurements in preoperative evaluation of candidates for pulmonary resection. J Thorac Cardiovasc Surg 1978;75:519–24.[Abstract]
10. Kearney DJ, Lee TH, Reilly JJ, DeCamp MM, Sugarbaker DJ. Assessment of operative risk in patients undergoing lung resection—importance of predicted pulmonary function. Chest 1994;105:753–9.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. Brunelli, A. Charloux, C. T. Bolliger, G. Rocco, J-P. Sculier, G. Varela, M. Licker, M. K. Ferguson, C. Faivre-Finn, R. M. Huber, et al. ERS/ESTS clinical guidelines on fitness for radical therapy in lung cancer patients (surgery and chemo-radiotherapy) Eur. Respir. J., July 1, 2009; 34(1): 17 - 41. [Abstract] [Full Text] [PDF]

				C. S. Reilly Can we accurately assess an individual's perioperative risk? Br. J. Anaesth., December 1, 2008; 101(6): 747 - 749. [Full Text] [PDF]

				W. K. Bernstein and S. Deshpande Preoperative Evaluation for Thoracic Surgery Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2008; 12(2): 109 - 121. [Abstract] [PDF]

				A. Brunelli, M. Refai, F. Xiume, M. Salati, R. Marasco, V. Sciarra, L. Socci, and A. Sabbatini Oxygen desaturation during maximal stair-climbing test and postoperative complications after major lung resections Eur. J. Cardiothorac. Surg., January 1, 2008; 33(1): 77 - 82. [Abstract] [Full Text] [PDF]

				G. L. Colice, S. Shafazand, J. P. Griffin, R. Keenan, and C. T. Bolliger Physiologic Evaluation of the Patient With Lung Cancer Being Considered for Resectional Surgery: ACCP Evidenced-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 161S - 177S. [Abstract] [Full Text] [PDF]

				A. Shaw Genetics of postoperative complications following thoracic surgery. Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2006; 10(4): 327 - 345. [Abstract] [PDF]

				A. Brunelli, M. Al Refai, M. Monteverde, A. Borri, M. Salati, and A. Fianchini Predictors of exercise oxygen desaturation following major lung resection Eur. J. Cardiothorac. Surg., July 1, 2003; 24(1): 145 - 148. [Abstract] [Full Text] [PDF]

				F. J. Algar, A. Alvarez, A. Salvatierra, C. Baamonde, J. L. Aranda, and F. J. Lopez-Pujol Predicting pulmonary complications after pneumonectomy for lung cancer Eur. J. Cardiothorac. Surg., February 1, 2003; 23(2): 201 - 208. [Abstract] [Full Text] [PDF]

				M. A. Beckles, S. G. Spiro, G. L. Colice, and R. M. Rudd The Physiologic Evaluation of Patients With Lung Cancer Being Considered for Resectional Surgery Chest, January 1, 2003; 123 (2009): 105S - 114S. [Abstract] [Full Text] [PDF]

				W H Allum, S M Griffin, A Watson, and D Colin-Jones Guidelines for the management of oesophageal and gastric cancer Gut, June 1, 2002; 50 (2007): v1 - v23. [Full Text] [PDF]

				J. R. Roberts, C. Eustis, R. Devore, D. Carbone, H. Choy, and D. Johnson Induction chemotherapy increases perioperative complications in patients undergoing resection for non-small cell lung cancer Ann. Thorac. Surg., September 1, 2001; 72(3): 885 - 888. [Abstract] [Full Text] [PDF]

				G. Varela, R. Cordovilla, M.F. Jimenez, and N. Novoa Utility of standardized exercise oximetry to predict cardiopulmonary morbidity after lung resection Eur. J. Cardiothorac. Surg., March 1, 2001; 19(3): 351 - 354. [Abstract] [Full Text] [PDF]

				British Thoracic Society, Society of Cardiothoracic Surgeons of Great Britai, and Ireland Working Party Guidelines on the selection of patients with lung cancer for surgery Thorax, February 1, 2001; 56(2): 89 - 108. [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Mathew Ninan K. Eric Sommers Rodney J. Landreneau James D. Luketich Peter F. Ferson Robert J. Keenan Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ninan, M. Articles by Keenan, R. J. Search for Related Content PubMed PubMed Citation Articles by Ninan, M. Articles by Keenan, R. J. Related Collections Related Article