| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anaesthesia and Pain Management, Alfred Hospital; and
Department of Psychology, Bethesda Hospital, Prahan, Victoria, Australia
Address correspondence to Dr. Paul S. Myles, Department of Anaesthesia and Pain Management, The Alfred, Commercial Rd., Prahan, Victoria, 3181 Australia. Address e-mail to p.myles{at}alfred.org.au
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendix 1References |
|---|
= 0.55, P < 0.0001). Discriminant construct validity was supported by comparing resultant QoR Scores in patients undergoing day-stay, minor, and major surgery (P = 0.008), as well as a negative correlation with duration of hospital stay ( = -0.20, P < 0.0001), and, using multivariate regression, demonstrating a significant negative relationship between QoR Score and female gender (P = 0.048) and older age (P = 0.041). There was also good interrater agreement ( = 0.55, P < 0.0001), test-retest reliability (median = 0.61, P < 0.0001), and internal consistency (
= 0.57 and 0.90, P < 0.0001). There was a significant difference between the groups of patients recovering from major and minor surgery (P < 0.001). This study demonstrates that the QoR Score has good validity, reliability, and clinical acceptability in patients undergoing many types of surgery. Implications: We set out to develop a patient-rated quality of recovery score (QoR) that could be used both as a measure of outcome in perioperative trials and for clinical audit. We first surveyed patients and staff to identify important aspects of recovery, then developed a nine-point QoR Score. This was then compared with other measures of postoperative outcome. We found that the QoR Score is a useful measure of recovery after anesthesia and surgery.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
1If you ask a diverse group of individuals this question, you will receive a variety of responses. For example, an anesthesiologist may suggest a smooth induction of anesthesia, stable hemodynamics, and rapid, pain-free recovery. Nursing staff may include a description of the patient's physical activity and emotional and social functions. The patient's immediate family and friends may have similar wishes or may only want to know that the patient has survived the procedure. Surgeons may demand optimal operating conditions, as well as a satisfied customer on postoperative review. Hospital and healthcare fund managers are under pressure to reduce costs and yet maintain quality of care.
Despite these many and varied opinions, the ultimate judge of a good anesthetic should be the patient undergoing the procedure (and the recovery process). Parallels to this have been discussed widely in other areas of healthcare, in particular in the assessment of management strategies for cancer and chronic disease (2–5). Thus, a range of quality of life measures have been developed to refocus the goals of medical care to patients' assessment of what constitutes a favorable outcome.
The aim of this study was to develop a patient-orientated quality of recovery score (QoR Score) after anesthesia and surgery, then to measure its validity and reliability in a large number of patients recovering from a diverse range of surgical procedures.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
|
These groups were chosen to represent a demographically broad spectrum of individuals. After answering the questionnaire, they were asked to add further comments that they considered to be relevant to a "good" postoperative recovery. Results are presented in Table 1. Most subjects rated most items quite highly, especially features that inquired about patient comfort and regaining independence and usual activity level, as well as patients' ability to communicate and sense support from hospital staff and family. When these ratings were compared among patients and their relatives, medical, and nursing staff, there were some notable differences. In general, medical staff rated items lower than nursing staff and patients/relatives. However, nursing staff rated some items higher than medical staff and patients/relatives (patient confusion, nausea, and moderate pain); they also rated some items significantly lower (vivid dreams and bad dreams). Patients/relatives also rated several items higher than nursing and medical staff: increased interest in work, self-hygiene, no diarrhea, enjoyment of food, ability to read, and ability to write.
This information was then collated, and, with the assistance of an experienced psychologist (JP), highly ranked items were incorporated into a nine-point instrument with a simple (ordinal) scoring system. The resultant QoR Score is presented in Figure 1.
|
Finally, the QoR Score was tested in daily clinical practice: 59 sequential inpatients were asked to rate their recovery on the morning after surgery. The correlation between their QoR Score and a global quality of recovery visual analog scale (VAS; see Appendix 1), a measure of convergent validity (3,6), was significant ( = 0.60, P < 0.0005). Testing of an expanded four-point scale on a further 31 patients found similar results for construct validity ( = 0.55, P < 0.0005), although this took longer for the patients to complete: 1.67 ± 0.68 vs 2.13 ± 0.39 min (P < 0.0005). Most (88%) patients interviewed on the day after surgery were able to complete the QoR Score themselves, with the remainder requiring assistance from a nurse interviewer.
Psychometric Evaluation
Patients >18 yr of age undergoing general anesthesia and surgery were recruited into this part of the study. Patients were excluded if they had poor English comprehension, a psychiatric disturbance that precluded complete cooperation, a known history of alcohol or drug dependence, any severe preexisting medical condition that limited objective assessment postoperatively (e.g., rheumatoid arthritis, unstable angina), or the presence of any life-threatening postoperative complication.
After ethics committee approval and informed consent, we studied two cohorts of patients. The first (Cohort A) were approached during their initial recovery after any of a broad range of surgical procedures. The second group (Cohort B) were recruited during the preoperative period and then observed for 6 wk postoperatively.
Cohort A.
A diverse group of postoperative patients (n = 389) was selected to test the QoR Score instrument (see Table 2). Patients were asked to rate themselves at a time when the investigators considered that they had made an initial satisfactory postoperative recovery. For day-stay patients, this was immediately before hospital discharge; for patients who had undergone minor surgery, this was the evening of or day after surgery; and for patients who had undergone major surgery, this was usually Days 2–4 after surgery.
|
Cohort B.
A sequential sample of patients undergoing either minor (n = 30) or major (n = 30) surgery was identified preoperatively to document the QoR Score changes over time. Measurements were performed at the following time points (telephone interview was used after discharge from hospital): preoperatively (baseline); on discharge from the recovery room; 4 h postoperatively; Postoperative Days 1–6; and Postoperative Weeks 1–6.
Validity
Validity is a measure of the extent to which scores accurately describe quality of recovery. This is difficult to verify because there are no accepted alternative methods for comparison. We used the following to evaluate validity.
Reliability
Reliability is a measure of consistency, the ability of the instrument to measure without error. We used the following to evaluate reliability:
. .
Statistical Analysis
The sample size selected for this project was chosen for convenience, given that power calculations cannot be readily made with correlational analysis. Based on our clinical experience, we expected that the large sample size selected would adequately represent our postoperative surgical population of interest.
Whether there were any rating differences in the 61-item questionnaire among the three groups (patients/relatives, nursing staff, medical staff) was investigated using a nonparametric Kruskall-Wallis analysis of variance. Each item is summarized by its standardized rank and median.
Nonparametric Spearman rank correlation () was used to measure associations (for validity and reliability testing). Internal consistency was measured using Cronbach's . The ability of the QoR Score to discriminate between groups (good recovery versus poor recovery) was measured in Cohort A using the Wilcoxon rank sum test. These groups were determined by averaging the VAS scores made (independently) by the surgical staff and researcher; the median of this pooled VAS score was 80. Recovery was considered good if the pooled VAS score was >80; recovery was considered poor if the pooled VAS score was 
80.
Data are presented as mean ± SD, median (10th–90th percentile) or number (%). The pattern of the postoperative QoR Score in Cohort B is described by median and 95% confidence intervals (11). QoR Scores for minor and major surgical patients and improvement over time were compared by using repeated-measured analysis of variance; the null hypothesis was rejected if P < 0.05. All analyses were performed using SPSS v4.0 (SPSS Inc., Chicago. IL).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
|
= 0.55, P < 0.0001. This correlation was consistent across all types of surgery ( = 0.51–0.74) and the extent of surgery (day-stay surgery, = 0.76; minor surgery, = 0.39; major surgery, = 0.58; all P < 0.0001).
Construct validity was tested by comparing the QoR Score of patients in Cohort A undergoing day-stay, minor, and major surgery. There was a significant decrease in QoR Score across these groups (Kruskall-Wallis analysis of variance P = 0.008). Similarly, using multivariate regression to adjust for imbalance between patient and surgical characteristics, we confirmed that older age (P = 0.041) and female gender (P = 0.048) were associated with a lower QoR Score. There was a significant negative correlation between the QoR Score and duration of hospital stay ( = -0.20, P < 0.0001). In Cohort B, there was a significant negative correlation between initial QoR Score and duration of surgery ( = -0.40, P = 0.002) and length of time spent in the recovery room ( = -0.32, P = 0.016).
Discriminant construct validity was determined by comparing patients who had undergone a good or poor postoperative recovery in Cohort A. The QoR Score differed significantly between these groups (good recovery 17[15–18] versus poor recovery 15[11–17]; P < 0.0001). The staff-rated QoR Score also differed significantly between these groups (17[15–18] vs 15[12–17]; P < 0.0001).
The pattern of change of the QoR Score in the groups of patients recovering from major and minor surgery (Cohort B) is presented in Figure 2. The median QoR Scores were significantly different between groups (P < 0.001).
|
= 0.55, P < 0.0001) and fair reliability for the staff-rated VAS ( = 0.29, P < 0.0001). There was also good agreement between the patient-rated QoR Score and the researcher-rated VAS ( = 0.69, P < 0.0001) and fair agreement for the staff-rated QoR Score ( = 0.26, P < 0.0001).
Internal consistency was tested using Cronbach's , which was 0.57 for Cohort A and 0.90 for Cohort B (both P < 0.0001). The split-half coefficient was 0.70 (P < 0.0001). Test-retest reliability was measured in Cohort B using scores obtained on the first 3 days of follow-up (median = 0.62, P < 0.0001). Repeat QoR Scores obtained over the 6-wk period were strongly correlated (multiple r = 0.89, P < 0.0001).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
There were some interesting differences in rating patterns among patients/relatives, nurses, and doctors. For patients/relatives, personal comfort, support, and being able to communicate were most important. For nurses, importance was placed on alertness, ability to communicate, and presence of several physical symptoms (particularly pain and emesis). Doctors tended to place significantly less importance than patents/relatives and nurses on depression, being in control, family support, and a number of "minor" physical symptoms.
Our validation methods remain open to criticism because there is no "gold standard" for quantifying a good quality postoperative recovery with which to compare our results. We therefore had to consider this outcome from a variety of perspectives, all of which support our conclusions. The evidence of construct validity was strong; the QoR Score was able to differentiate known determinants of postoperative recovery. We were able to demonstrate a negative association between the QoR Score and duration of surgery, time spent in the recovery room, and duration of hospital stay. The score clearly distinguished patient groups who had undergone either minor or major surgery, with the greatest difference existing over the first 3 postoperative days. Previously published recommendations state that, for an instrument to detect changes within individuals, reliability should exceed 0.70 (12,13). The QoR Score did not quite achieve this level of reliability in all respects. Nevertheless, the QoR Score should provide reliable assessment for group measurement and/or comparison.
Kirshner and Guyatt (13) argue that appraisal of health status instruments should be dictated by the instrument's intended application. Because we expect that the QoR score will most often be used as an outcome measure in clinical trials and as an audit tool (i.e., evaluation), the most relevant measure of reliability is responsiveness (13). Kirshner and Guyatt (13) defined this as the ability of an instrument to detect clinically significant changes in health status. Responsiveness can be calculated as the square root of twice the mean square error; in our study, this results in a value of approximately 1.0. This suggests that a clinical trial would require approximately 20 patients per group to detect a clinically significant improvement in the quality of a postoperative recovery (14). Similarly, Kazis et al. (15) introduced the concept of effect size to interpret changes in health status measurements: an effect size represents a standardized measure of change calculated as the mean change divided by the pretreatment standard deviation. In our study, the QoR Score has an effect size of approximately 2.0. This suggests a very strong ability to detect a clinically important change in quality of recovery.
The QoR Score is a summary measure, a simple number. Some authors have criticized using such methods (3), but we contend that it is a simple, valid measurement. This is supported by Katz et al. (16), who compared short and longer health status instruments in patients recovering from hip surgery and found that they had similar sensitivity to clinical change. We had a high rate of participation (99.6%) and successful completion (88%), and there were minimal missing data (a surrogate marker of clinical applicability). This should enhance the scale's clinical acceptability and widespread use, both as an audit tool (quality of care) and as a measure of outcome in perioperative clinical trials.
We recognize that results from our original group of patients (Cohort A) were prone to selection bias, as patients were approached after undergoing an initially satisfactory period of postoperative recovery. This led to higher QoR Scores overall and would not be representative of the typical surgical population. We intentionally chose to study this selected population because we required greater cooperation from them and their treating staff. We also observed a sequential cohort through the whole perioperative period (Cohort B); this group would more accurately describe the pattern of the changing QoR Score over time. This pattern is consistent with the expected early postoperative course of most patients: those assessed after minor surgery have (mostly) recovered within 24 h, and those assessed after major surgery require at least 3 days to recover.
Previous attempts at measuring aspects of recovery after surgery and anesthesia have focused on: physiological end points, such as time to awakening, coordination, and ambulation (17–20); presence and extent of a variety of adverse events, such as pain, emesis, major organ failure, and mortality (7–10,18,21,22); economic end points such as duration of hospital stay, readmission rates, and cost of treatment (23,24); and psychosocial status, by measuring mood and patient satisfaction (25–27). Although all of these outcomes are probably important, the value that patients place on each may vary markedly at different times during their recovery (1,28). Once patients realize that they have survived an operation, their subjective experience may change as their expectations increase. Similarly, patients who have had a previous experience of surgery and anesthesia may have high (or low) expectations of their postoperative recovery period (1). Because advances have been made to reduce mortality and major morbidity associated with anesthesia (22), efforts should now be focused on improvements in quality.
The range of quality of life instruments available is not suitable for our purposes, as they are often disease-specific, have a different focus, or require expertise in their execution. Some methods used to measure mood and patient satisfaction have not been validated in the postoperative population (29). An exception to this is a recent study by Dexter et al. (27), who developed a valid and reliable measure of patient satisfaction in 86 patients after monitored anesthesia care. It is uncertain how useful this instrument will be in the broader range of patients recovering from general anesthesia. Our score encompasses a range of patient conditions and was tested after a wide variety of surgical procedures.
Comparison of the quality of recovery should be an accepted outcome measure in many anesthetic, surgical, and other perioperative trials. The relationship with currently used measures of patient satisfaction or quality of life can also be explored. Given the increasing economic constraints on healthcare, it is imperative that changes in practice or purported advances in care be tested with consideration of the patient's perspective. The QoR Score can be used to address this deficiency.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
.
|
|
Acknowledgments |
|---|
We thank Ms. Melanie Hendrata for her expert assistance and comments regarding the application of the QoR Score during our routine postoperative quality assurance activities and Dr. Tony Weeks for his thoughtful and constructive criticisms of the manuscript. We also thank our colleagues (nursing staff, residents, and anesthesiologists) for their interest and cooperation.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Tufanogullari, P. F. White, M. P. Peixoto, D. Kianpour, T. Lacour, J. Griffin, G. Skrivanek, A. Macaluso, M. Shah, and D. A. Provost Dexmedetomidine Infusion During Laparoscopic Bariatric Surgery: The Effect on Recovery Outcome Variables Anesth. Analg., June 1, 2008; 106(6): 1741 - 1748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. A. Caljouw, M. van Beuzekom, and F. Boer Patient's satisfaction with perioperative care: development, validation, and application of a questionnaire Br. J. Anaesth., May 1, 2008; 100(5): 637 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Chung and N. Assmann Car Accidents After Ambulatory Surgery in Patients Without an Escort Anesth. Analg., March 1, 2008; 106(3): 817 - 820. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Sun, O. Sacan, P. F. White, J. Coleman, R. J. Rohrich, and J. M. Kenkel Perioperative Versus Postoperative Celecoxib on Patient Outcomes After Major Plastic Surgery Procedures Anesth. Analg., March 1, 2008; 106(3): 950 - 958. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Bost, B. A. Williams, M. T. Bottegal, Q. Dang, and D. M. Rubio The 8-Item Short-Form Health Survey and the Physical Comfort Composite Score of the Quality of Recovery 40-Item Scale Provide the Most Responsive Assessments of Pain, Physical Function, and Mental Function During the First 4 Days After Ambulatory Knee Surgery with Regional Anesthesia Anesth. Analg., December 1, 2007; 105(6): 1693 - 1700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Bryson, C. Thompson, S. Gagne, L. Byford, J. Penning, and M. Kattan The addition of adrenaline to thoracic epidural meperidine does not improve analgesia following thoracotomy: [L'ajout d'adrenaline a la meperidine peridurale thoracique n'ameliore pas l'analgesie a la suite d'une thoracotomie] Can J Anesth, November 1, 2007; 54(11): 882 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Liu and C. L. Wu The Effect of Analgesic Technique on Postoperative Patient-Reported Outcomes Including Analgesia: A Systematic Review Anesth. Analg., September 1, 2007; 105(3): 789 - 808. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. L. Bryson, R. MacNeil, L. M. Jeyaraj, and O. P. Rosaeg Small dose spinal bupivacaine for Cesarean delivery does not reduce hypotension but accelerates motor recovery: [Une faible dose de bupivacaine intrathecale lors de l'accouchement par cesarienne ne reduit pas l'hypotension, mais accelere la recuperation motrice] Can J Anesth, July 1, 2007; 54(7): 531 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Herrera, J. Wong, and F. Chung A Systematic Review of Postoperative Recovery Outcomes Measurements After Ambulatory Surgery Anesth. Analg., July 1, 2007; 105(1): 63 - 69. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. White, O. Sacan, B. Tufanogullari, M. Eng, N. Nuangchamnong, and B. Ogunnaike Effect of short-term postoperative celecoxib administration on patient outcome after outpatient laparoscopic surgery: [Effet de l'administration postoperatoire a court terme de celecoxib sur l'evolution des patients apres une chirurgie par laparoscopie sans hospitalisation] Can J Anesth, May 1, 2007; 54(5): 342 - 348. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. MacNeil, G. Bryson, L. Jeyaraj, and O. Rosaeg 26337 - MINI-DOSE SPINAL BUPIVACAINE FOR CASEAREAN SECTION. Can J Anesth, June 1, 2006; 53(suppl_1): 26337 - 26337. [Full Text] [PDF]
|
|
|
|
|
|
P. F. White, J. Tang, M. A. Hamza, B. Ogunnaike, M. Lo, R. H. Wender, R. Naruse, A. Sloninsky, R. Kariger, S. Cunneen, et al. The use of oral granisetron versus intravenous ondansetron for antiemetic prophylaxis in patients undergoing laparoscopic surgery: the effect on emetic symptoms and quality of recovery. Anesth. Analg., May 1, 2006; 102(5): 1387 - 1393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Turan, G. Kaya, B. Karamanlioglu, Z. Pamukcu, and C. C. Apfel Effect of oral gabapentin on postoperative epidural analgesia Br. J. Anaesth., February 1, 2006; 96(2): 242 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Turan, P. F. White, B. Karamanlioglu, D. Memis, M. Tasdogan, Z. Pamukcu, and E. Yavuz Gabapentin: An Alternative to the Cyclooxygenase-2 Inhibitors for Perioperative Pain Management Anesth. Analg., January 1, 2006; 102(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. G. Wakeling, M. R. McFall, C. S. Jenkins, W. G. A. Woods, W. F. A. Miles, G. R. Barclay, and S. C. Fleming Intraoperative oesophageal Doppler guided fluid management shortens postoperative hospital stay after major bowel surgery Br. J. Anaesth., November 1, 2005; 95(5): 634 - 642. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Morris, M. Acheson, M. Reeves, and P. S. Myles Effect of clonidine pre-medication on propofol requirements during lower extremity vascular surgery: a randomized controlled trial Br. J. Anaesth., August 1, 2005; 95(2): 183 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Fung, M. Cohen, S. Stewart, and A. Davies Can the Iowa Satisfaction with Anesthesia Scale Be Used to Measure Patient Satisfaction with Cataract Care Under Topical Local Anesthesia and Monitored Sedation at a Community Hospital? Anesth. Analg., June 1, 2005; 100(6): 1637 - 1643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Apfelbaum, T. J. Gan, S. Zhao, D. B. Hanna, and C. Chen Reliability and Validity of the Perioperative Opioid-Related Symptom Distress Scale Anesth. Analg., September 1, 2004; 99(3): 699 - 709. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ma, J. Tang, P. F. White, A. Zaentz, R. H. Wender, A. Sloninsky, R. Naruse, R. Kariger, R. Quon, D. Wood, et al. Perioperative Rofecoxib Improves Early Recovery After Outpatient Herniorrhaphy Anesth. Analg., April 1, 2004; 98(4): 970 - 975. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. F. K. Kwok, J. Lim, M. T.V Chan, T. Gin, and W. K.Y. Chiu Preoperative Ketamine Improves Postoperative Analgesia After Gynecologic Laparoscopic Surgery Anesth. Analg., April 1, 2004; 98(4): 1044 - 1049. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Recart, I. Gasanova, P. F. White, T. Thomas, B. Ogunnaike, M. Hamza, and A. Wang The Effect of Cerebral Monitoring on Recovery After General Anesthesia: A Comparison of the Auditory Evoked Potential and Bispectral Index Devices with Standard Clinical Practice Anesth. Analg., December 1, 2003; 97(6): 1667 - 1674. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Wu, M. Naqibuddin, A. J. Rowlingson, S. A. Lietman, R. M. Jermyn, and L. A. Fleisher The Effect of Pain on Health-Related Quality of Life in the Immediate Postoperative Period Anesth. Analg., October 1, 2003; 97(4): 1078 - 1085. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Lee, P. T. Chui, and T. Gin Educating Patients About Anesthesia: A Systematic Review of Randomized Controlled Trials of Media-Based Interventions Anesth. Analg., May 1, 2003; 96(5): 1424 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. J. Eberhart, R. Mayer, O. Betz, S. Tsolakidis, W. Hilpert, A. M. Morin, G. Geldner, H. Wulf, and W. Seeling Ginger Does Not Prevent Postoperative Nausea and Vomiting After Laparoscopic Surgery Anesth. Analg., April 1, 2003; 96(4): 995 - 998. [Abstract] [Full Text] |
