|
|
|
|
 |
 |
|
|
|
|
|
|
|
Issues in Pulmonary Anesthesiology
"Airway Obstruction and Perioperative Complications in Smokers Undergoing Abdominal Surgery"
1Warner, DO, Warner, MA, Offord, KP, Schroeder, DR, Maxson, P, Scanlon, PD. Airway Obstruction and Perioperative Complications in Smokers Undergoing Abdominal Surgery Anesthesiology 1999 February; 90(2): 372-279
This study recognizes the common use of preoperative spirometry in the assessment of certain patient subpopulations, particularly subpopulations with pre-existing pulmonary compromise.
The question has to do is whether these preoperative spirometric analyses might be useful in predicting a perioperative complications in a group considered probably at risk, smokers.
To focus this research, a number of specific criteria were identified in the common surgical procedure applying toall patients was abdominal surgery.
Within the patient group undergoing abdominal surgery, a number of specific pulmonary-related issues were considered.
These included:
FEV1 < 40% of the predicted value
the FEV1/FVC ratio was < the lower limit of the probable normal value
Smoking history > 20 pack-years and
Age> 35.
The control group was matched in terms of gender, surgical site, smoking history, and age but did not exhibit airway obstruction.
The number of patients in each group was 135.
Perioperative complications, defined as those occurring within 30 days following surgery, were assessed by medical record review.
The problem has to do with the possibility if not likelihood that patients with COPD may experience postoperative complications more often than patients who have normal pulmonary function.
Avoidance of prolonged postoperative mechanical ventilatory support, a serious postoperative complications, is highly desirable.
Concern about surgery on patients with significant COPD has been reflected in an effort to optimize preoperatively pulmonary function, sometimes avoiding general anesthesia, or even denying surgery to the station group.
Studies that have reached the above basic conclusion had been themselves subject to experimental limitations.
For example, inadequate control of factors other than airway obstruction such as surgical site, other pathophysiological conditions that add to risk, and smoking history has been a problem.
Also, as one compares different studies, there were differences in definitions about what constitutes postoperative pulmonary complications, what defines chronic lung disease, etc.
Some complications alleged to be pulmonary nature (e.g. fever) could have been due to other nonpulmonary causes.
If one uses atelectasis based on chest x-ray as evidence of complication, one makes a questionable assumption about the clinical relevance or significance of this finding.
The authors also note that preoperative lung disease severity in many studies is not described or turns out to be described as fairly mild.
Finally, depending on when previous studies were performed, there may be an imporant disparity between the then state-of-the-art anesthetic, surgical, and postsurgical care compared to the present standard of care.
The general observation is that patients with COPD usually tolerate surgery fairly well and if one uses postoperative ventilatory failure as an endpoint, about 5% of patients will exhibit this complication.
Generally, preoperative pulmonary function testing and its relationship to perioperative frequency/severity of complication has been difficult to determine definitively
Therefore, the underlying issue of significance of airway obstruction in terms of a separate identifiable risk factor for perioperative pulmonary complications in this patient group becomes an important research question.
The authors make an effort to control for many of the experimental difficulties by this focus study in which abdominal surgery is considered along with significant smoking history.
The hypothesis is that perioperative respiratory function deficit will translate to an increase likelihood of prolonged postoperative intubation (defined as > 24 hours) in patients with severe airway obstruction.
Furthermore, this result would occur despite clinical efforts in place to prevent this result.
So, restating, medical records of 2 cohorts of patients, both exhibiting comparable smoking history, both gender and age matched , but differing in terms of spirometric identification of airway obstruction were examined.
The primary matching is for smoking history and surgical site.
The information concerning pulmonary function was derived from the database which all functional test results.
Associated with this information was questionnaire-based data concerning pulmonary history, including smoking status.
Patient history information was obtained during an interview by "certified pulmonary function technologist".
The relevance of this point has to do with the presumed increased likelihood of correct information assessed by questionnaire if a certified pulmonary specialist is involved.
The correctness of the information becomes central to the remaining study aspects since this information is used in the development of the cohorts.
Severe airway obstruction was defined in terms of FEV1 < 40% of predicted normal, ratio of FEV1/FVC < lower limit of predicted normal, smoking history > 20 pack-years and age > 35.
Now a couple things: the criteria includes COPD patients without distinguishing between specific COPD causes, such as bronchitis, asthma, or emphysema.
Also, the cutoff of 40% in FEV1, presumably not entirely arbitrary on one hand, probably has some intrinsic uncertainty in the value, as it compares against a predicted norm.
Similarly, the FEV1/FVC ratio also has some degree of uncertainty and as again it references a predicted norm.
These concerns about uncertainty are independent of concerns having to do with the repeatability or the variance of the individual FEV1 value, e.g. what is the uncertainty in the FEV1 number in itself when the determination is repeated in a patient.
Nevertheless, these are the bases for selection and these were consistently applied.
Following identification of the two groups, differing in terms of presence or absence of airway obstruction determined by spirometry, cross-referencing was implemented to identify patients who had undergone abdominal surgery (ICD-9 codes).
Laparoscopic procedures were excluded.
Each patient with airway obstruction was then matched with a patient without airway obstruction -- matching including gender, incision site (lower vs. upper abdominal site), pack-year smoking history, and age.
Additional information beyond pulmonary function and data included ASA number, presence of comorbid medical condition(s), and intraoperative information including type induration of anesthesia and characteristics of postoperative analgesia.
Surgical incision site was noted in the distinction between upper and lower abdominal surgery was as follows: incisions extending cephalad and caudad to the umbilicus were categorized as upper incisions.
Outcomes were assessed as follows:
Pulmonary embolus identification was based on either pulmonary angiography or high probability following analysis of ventilation-perfusion scans.
Pneumonia diagnosis was dependent on the presence of the new infiltrate observed on chest x-ray along with the presence of fever, leukocytosis, and positive sputum Gram stain.
Probable pneumonia used the same criteria with the exception of positive Gram stain data.
Prolonged tracheal intubation was determined if either it was not possible to extubate the patient within 24 hours following surgery or reintubation was required along with mechanical ventilation in a patient previously extubated for more than 15 minutes.
Bronchospasm was considered present if both bronchospasm, wheezing, increased air pressure, or extended expiration duration was noted and resolution of the condition required bronchodilator treatment beyond the reinstitution of the patient' s normal preoperative medication protocol.
Prolonged admission to an ICU (> 96 hours), ICU readmission following discharge, death, myocardial infarction, CNS morbidity, renal dysfunction, sepsis, and biliary dysfunction were also noted.
The time elapsed between the pulmonary test data and surgery was about nine days for patients with and without airway obstruction (10 and 8 days).
The types of surgical procedures were considered similar for the two groups. The specific data may be found in the original article.
The FEV1 was about 0.9 +/- 0.2 and 2.9 +/- 0.6 L for patients with and without obstruction. These values corresponded to about 30% +/- 6% and 92% +/- 10% of predicted values.
The two groups were matched for gender, but patients with airway obstruction is turned out to be slightly (but statistically significantly) older than those without obstruction.
In terms of age matching 95% of pairs were +/- 5 years.
In terms of pack-year smoking history, matching occurred within +/- 10 pack-years for about 85% of matched pairs.
In terms of smoking history, there tended to be a greater proportion of patients without airway obstruction that were current smokers, compared to those with spirometric-determined airway obstruction.
Those patients with airway obstruction also tended to have a higher ASA number.
There is no apparent difference between the groups in terms of the percentage of patients receiving regional anesthesia and those that did not or with respect to anesthesia duration or with respect to the use of postoperative epidural analgesia.
The details of these descriptions are found in the original article.
About 16% of patients (22) without airway obstruction and about 25% (34) of patients with airway obstruction had at least a single postoperative complication.
Following additional statistical considerations, a subset of outcomes were evaluated for intercohort comparison; these were the outcomes that occurred with sufficient frequency to allow reasonable statistical analysis..
These outcomes included bronchospasm, prolonged intubation, total intensive care stay > 96 hours and ICU readmission.
The results indicated that the presence of airway obstruction was a significant risk factor for the bronchospasm outcome.
Patients with airway obstruction tended to be more likely to be admitted to the ICU immediately following surgery than those without airway obstruction, although the duration of postoperative ICU stay didn't differ between groups. Addition results can be found in the original paper.
One conclusion of the article is that a significant number of predominantly elderly patients had complications after abdominal surgery and that at least for those complications that occurred with significant frequency to allow reasonable analysis, airway obstruction, as defined by spirometric evaluation, predicted only bronchospasm.
The presence of airway obstruction did not increase the frequency of prolonged endotracheal intubation.
This conclusion, concerning bronchospasm, does not mean that spirometry could not in principal predicted other complications but only that the frequency of these other complications was too low to allow assessment.
Therefore, a larger study might have revealed (or not) other associations.
The authors of the study notes that there had been three large prospective studies comparing pulmonary outcome in patients with or without lung disease.
These studies did not provide diagnostic COPD criteria nor pulmonary function data and accordingly despite their size may have relatively limited value in resolving the central questions.
However, despite these limitations the results of the other studies were as follows:
Wightman2 noted in a series of 785 patients undergoing nonthoracic surgery, there were postoperative pulmonary complications at a frequency of 8% in the group without chronic respiratory disease; moreover, 26 percent of patients with chronic respiratory disease had postoperative pulmonary complications.
An even larger study by Pedersen3 involving 7000 patients undergoing non thoracic surgery found that postoperative complications obtained in 4% of patients without lung disease compared to 12% of patients was COPD.
In that study, statistical analysis taking into consideration other risk factors suggested that COPD was an important risk factor for postoperative pulmonary complications, corresponding to an approximate threefold increased risk.
The largest study, a multicenter study involving 17,000 patients receiving general anesthesia for all types of surgery, indicated that again COPD predicted significant, severe pulmonary outcome, again a threefold increased in risk compared to non-COPD patients.4
Despite the size of the studies, critical research methodological controls probably prevent definitive conclusions.
As pointed out by the current authors, Lawrence5 in reviewing the 22 research studies available for 1989 noted that they all appeared to contain significant methodological flaws preventing valid conclusions concerning possible relationships between airway obstruction (spirometric analysis) and pulmonary complications.
These research problems ranged from selection bias, lack of inclusion of patients healthy lungs, lack of appropriate management of other risk factors, lack of blinding to inappropriate or non-consistent outcome definition.
The present study appears to be a step forward in critical assessment in part by focusing on a higher risk patient population, smokers, and by focusing on the influence of airway obstruction.
Use of FEV1 allowed the exclusion of restrictive lung disease and the matching of clinical profiles was central to the study.
Also, well-defined, clinically relevant outcome events were chosen and were chosen such that they could be readily extracted from medical record review.
These authors also note certain cautions that should be applied in evaluating their study.
One caution was that patients without spirometric-determined airway obstruction were not necessarily devoid of respiratory disease-all had significant smoking history which translated to sufficient concern such that preoperative pulmonary function analysis was obtained.
Accordingly, the study did not compare patients with COPD in patients with healthy lungs.
Perioperative complications are typically more likely associated with COPD/respiratory disease patients compared to patients without these syndromes.
Another concern is that the presence of airway obstruction itself could influence perioperative management which would then in turn affect complication frequency.
The airway obstruction patients might be in principal more likely to be ICU admits right after surgery and there might receive more substantial respiratory therapy intervention.
Because of the particular methods involved in the study, study results probably represent outcomes achieved when preoperative pulmonary function was optimized and when postoperative care was applied commensurate with clinical discretion.
However, the study could not determine whether airway obstruction presence would affect perioperative risk if perioperative respiratory therapy will standardize between the two groups.
The authors of the present work concluded that in patients with a significant history of smoking, who undergo abdominal surgery and are treated in accord with current clinical practice, perioperative bronchospasm is more likely develop in those patients with significant airway obstruction as assessed by preoperative spirometric evaluation.
Furthermore, preoperative airway obstruction did not predict prolonged intubation when taking into account other factors including age, smoking history, and surgical incisions site.
|
|
|
|
|
1Warner, DO, Warner, MA, Offord, KP, Schroeder, DR, Maxson, P, Scanlon, PD. Airway Obstruction and Perioperative Complications in Smokers Undergoing Abdominal Surgery Anesthesiology 1999 February; 90(2): 372-279--articles below, second sourced from ref. 1.
2Wightman, JAK: A prospective survey of the incidence of postoperative pulmonary complications. Br J. Surg 1968; 55:85-91
3Pedersen, T, Eliasen, K, Henriksen, E: A prospective study of risk factors and cardiopulmonary morbidity. Acta Anaesthesiol Scand 1990; 34: 144-55.
4 Forrest, JB, Rehder, K, Cahalan, MK, Goldsmith, CH: Multicenter study of general anesthesia. III. Predictors of severe perioperative adverse outcomes. Anesthesiology 1992; 76: 3-15.
5Lawrence, VA, Page, CP, Harris GD: Preoperative spirometry before abdominal operations. A critical appraisal of its predictive value. Arch Intern Med 1989;149: 280-5.
|
|
|