Subxiphoid versus Intercostal Chest Tubes

Abstract

Chest tubes are one cause of pain after cardiac surgery. In a prospective, randomized study, we investigated the effects of the position of chest tubes on acute postoperative pain and pulmonary morbidities in patients who underwent coronary artery bypass grafting.

From June through December 2010, 40 patients who underwent elective coronary artery bypass grafting were enrolled in the study. We investigated 2 randomized groups of patients: Group 1 (n–20) had a left chest tube inserted through the midline inferior to the xiphoid process (subxiphoid approach), and Group 2 (n–20) had a left chest tube inserted through the 6th intercostal space along the anterior axillary line (intercostal approach). We compared the results with respect to postoperative pain, the need for analgesic agents, chesttube drainage, pulmonary morbidities, and duration of hospitalization.

The intensity of postoperative pain was similar between the groups. The cumulative doses of analgesic agents, incidence of pulmonary morbidities, and duration of hospitalization were also similar. Pleural effusion and atelectasis were each diagnosed in 3 patients in Group 1 (15%) and 1 patient in Group 2 (5%) (both P=0.68). Two of the patients in Group 1 required drainage of the pleural effusion.

In our study, we found that the subxiphoid and intercostal approaches for chest-tube placement yielded similar clinical outcomes.

In patients who undergo coronary artery bypass grafting (CABG), iatrogenic injuries to the pleura and the harvesting of the left internal thoracic artery (ITA) necessitate the placement of chest tubes to prevent hemothorax or pneumotho-rax after surgery. Tube placement can be performed through a subxiphoid or intercostal approach. Irritation of the pleura and intercostal nerves by friction from the chest tubes can cause postoperative pain and pulmonary morbidities.^1–4 Consequently, additional analgesic agents, chest physiotherapy for pulmonary sequelae, medical or surgical treatment of pleural fluid collection, and longer periods of hospitalization may be required.

Pain after CABG can also result from sternal or rib retraction, musculoskeletal trauma, sternal dehiscence or nonunion, harvesting of the ITA by means of cauterization, fractured ribs, intercostal nerve injury, and the use of sternal wires.¹ Specifically, pain associated with chest tubes can diminish pulmonary functions because of hypoventilation and atelectasis, particularly in high-risk CABG patients who have pulmonary disease.^2,3 Among reports of the impact of different routes for tube insertion,^4,5 one study⁴ indicated that the subxiphoid route leads to significantly less impairment of pulmonary function and less subjective pain than does intercostal insertion. Conversely, in our surgical practice, we typically insert chest tubes into the left hemitho-rax through the intercostal approach, for better drainage of blood and air. Because there have been relatively few studies of this issue, we compared the subxiphoid and intercostal routes of chest-tube insertion with respect to postoperative pain, analgesic requirements, and postoperative pulmonary morbidities in patients who underwent CABG with harvesting of the left ITA.

Patients and Methods

In this prospective study, we screened 162 patients who were scheduled to undergo elective CABG at Istanbul Sema Hospital from June through December 2010. Of that number, 40 patients were enrolled in the study. All operations involved opening the pleura and harvesting the left ITA. The patients were randomized into 2 groups by means of a computer-generated sequence. The 20 Group 1 patients were to have a left chest tube inserted through the midline, inferior to the xiphoid process (the subxiphoid approach). The 20 Group 2 patients were to have a left chest tube inserted through the 6th intercostal space along the anterior axillary line (the intercostal approach). The institutional board of the hospital reviewed and approved the study protocol. Each patient provided written informed consent for the surgery and follow-up consultations.

Exclusion Criteria. We excluded patients from the study population if they had left ventricular ejection fractions of less than 0.50, obstructive lung disease, renal insufficiency (preoperative creatinine level, >1.5 mg/dL), histories of alcohol or drug abuse, or neurologic dysfunction. In addition, we excluded patients who were taking nonsteroidal anti-inflammatory drugs (NSAIDs) on a chronic basis or tranquilizers before surgery, because these medications might have affected the evaluation of postoperative analgesia. During preoperative evaluation on the day before surgery, we explained to each patient the surgical procedure, the plans for postoperative care, and a visual analogue scale (VAS) and verbal rating scale (VRS) for pain. Patients who were unable to express themselves or who were unable to cooperate with the scale questionnaires were excluded from the study.

The VAS score (range, 0–10) ranged from “no pain” to “worst pain imaginable” or “pain as bad as it could be.” The VRS score (range, 0–5) was used to evaluate the severity of pain after deep inspiration, as follows: 0 = no pain at rest and while moving in bed; 1 = pain when coughing but without pain during deep breathing; 2 = mild to moderate pain during deep breathing but without pain while at rest; 3 = mild pain while at rest; 4 = moderate pain while at rest; and 5 = severe pain while at rest.

Anesthesia Technique

All patients were given 0.1 mg/kg of diazepam intramuscularly 30 min before surgery. General anesthesia was induced with use of midazolam (0.1 mg/kg), fentanyl (5 μg/kg), vecuronium (0.1 mg/kg) and propo-fol (0.5 mg/kg). Anesthesia was maintained with use of fentanyl, midazolam, sevoflurane, and a 50% mixture of air and oxygen. Additional bolus doses of fentanyl (100–200 μg) were given during the operations if needed. Neither fentanyl nor morphine was administered during the postoperative course.

Surgical Technique

The same surgeon performed all operations. All patients underwent median sternotomy and harvesting of the left ITA. The ITA was harvested with a pedicle in standard fashion, with use of cauterization and a chest retractor on one side of the sternum after wide opening of the left pleura. Standard on-pump CABG was performed with membrane oxygenation and moderate hypothermia. Myocardial preservation was achieved through the intermittent administration of isothermic blood cardioplegic solution.

In each patient, 2 soft, 32F chest tubes were used after surgery to drain blood and air from the chest cavity (Fig. 1). In the Group 1 patients, chest tubes were inserted through the subxiphoid area. An angled tube was placed in the left hemithorax, and the tip of this tube was placed in the left costophrenic sinus above the diaphragm. A straight tube was placed in the anterior mediastinum. In the Group 2 patients, a straight tube was inserted in the left hemithorax through the 6th intercostal space along the mid-axillary line, and the tip of this tube was directed toward the apex of the left lung. Another straight tube was passed through the subxiphoid area and placed in the anterior mediastinum. All tubes were connected to an underwater drainage system. Chest radiographs confirmed the placement and position of the tubes.

Fig. 1 Illustrations of the A) subxiphoid and B) intercostal routes of chest-tube placement.

Before drain removal, all patients were put into an upright position early on the first postoperative day, to enable the evacuation of residual fluid in the pericardial or pleural cavity. The chest tubes were removed 48 hours after postoperative extubation. The sternotomy was closed with 8 separate steel wires, and the skin incision was closed. The patients were then transferred to the intensive care unit and extubated by the same anesthesiologist.

Postoperative Analgesia Protocol

Postoperatively, NSAIDs were administered as analgesic agents.⁶ All patients were routinely given 1 mg/kg of diclofenac sodium every 12 hours during the first 48 hours after postoperative extubation. If a patient's VAS pain score was from 5 through 8 or the VRS pain score was 3 or 4, 1 mg/kg of tramadol was delivered intravenously. If pain persisted, the VAS score was above 8, or the VRS score was above 4, 1 mg/kg of meperidine was injected intramuscularly. The same surgeon recorded all pain scores and consumption of analgesics. As the patients rested or coughed, the effects of analgesia were evaluated in accordance with VAS and VRS scores at 1, 3, 6, 12, 18, 24, and 48 hours after postoperative extubation, after which time the chest tubes were removed.

Pulmonary Morbidities

Postoperative pulmonary morbidities were recorded, including atelectasis, pleural effusion, and the amount of drainage from the chest tubes. Also recorded were breaths per minute and the results of blood gas analysis, including Po₂, Pco₂, and oxygen saturation levels. Blood gas samples were collected 6 times daily. Postoperative chest-tube drainage in both groups was recorded hourly for 2 days. To evaluate intrapleural fluid retention, atelectasis, and diaphragmatic elevation, chest radiography was performed on each of the 2 days after surgery. Chest radiography was also used to detect residual pleural fluid on postoperative day 4, again upon each patient's discharge from the hospital, and finally at the 1-month follow-up examination. Pleural drainage was performed by the same surgeon.

Statistical Analysis

Initial sample-size estimation showed that approximately 18 patients were needed in each group to detect a clinically relevant reduction in pain level by 25%, with a power of 0.80 and a significance level of 5%. Statistical analysis was performed with use of SPSS for Windows, version 15.0 (IBM Corporation; Armonk, NY). Demographic characteristics and perioperative data were expressed as mean ± SD or as number of patients. Pain-intensity data were expressed as mean ± SE. Variables were tested for normal distribution by means of the Kol-mogorov-Smirnov test. The Student t, Mann-Whitney U, and χ² tests were used to compare the means of normally distributed data. For repeated measures, analysis of variance testing was performed. The Bonferroni correction was performed to compensate for the possible effects of repeated testing. Sphericity was evaluated by means of Mauchly's test of sphericity. A P value <0.05 was considered to be statistically significant.

Results

The groups did not differ significantly from each other with regard to age, sex, weight, height, body mass index, left ventricular function, and preoperative comorbidities (Table I). Perioperative data were similar with regard to durations of cardiopulmonary bypass, aortic cross-clamping, surgery, mechanical ventilation, and hospitalization (Table II). There was no significant difference between the groups in intraoperative fentanyl or postoperative analgesic consumption.

TABLE I. Comparison of Demographic and Clinical Characteristics between Groups

TABLE II. Comparison of Perioperative Data between Groups

Figure 2 shows the patients' postoperative VAS scores, and Figure 3 shows the VRS scores. At all of the established times of measurement, the mean VAS and VRS pain scores were similar in both groups.

Fig. 2 Distribution of mean visual analogue scale (VAS) scores for pain A) at rest and B) during coughing in Group 1 patients (subxiphoid chest tubes) and Group 2 patients (intercostal chest tubes).

Fig. 3 Distribution of mean verbal rating scale scores for pain after postoperative extubation in Group 1 patients (subxiphoid chest tubes) and Group 2 patients (intercostal chest tubes).

Postoperative pulmonary sequelae in the subxiphoid and intercostal approaches included pleural effusion (3 vs 1 patients, respectively; P=0.68) and atelectasis (3 vs 1 patients, respectively; P=0.68). Of the 3 patients in Group 1 who developed pleural effusion, 2 underwent in-hospital drainage. The chest radiographs of these patients were normal at the 1-month follow-up examination. There were no significant differences between the groups in respiratory rates or values yielded by blood gas analysis (Table III).

TABLE III. Comparison of Postoperative Pulmonary Measunents between Groups

Discussion

Our study showed that either the intercostal or the subxiphoid route for chest-tube insertion can be used in CABG patients without an associated pulmonary morbidity. There were no significant differences between the techniques in pulmonary function in the early postoperative period. The patients in Group 1 had higher percentages of pleural effusion and of atelectasis (both 15% vs 5%) and a consequent need for pleural drainage. These differences were not statistically significant (both P=0.68), and a larger sample size is needed to reach a definitive conclusion. In addition, no significant difference was observed in the comparative intensity of postoperative pain.

Pulmonary morbidities that occur after CABG include atelectasis, hypoxemia, pleural effusion, hemotho-rax, pneumothorax, and increased bronchial secretions or mucus retention. Sternotomy and pleurotomy can damage pulmonary mechanics and lead to deterioration in pulmonary function.³ Harvesting of the ITA can inhibit pulmonary function in the early postoperative period, especially when pleural integrity is disturbed.⁵ Inadequate analgesia and chest physiotherapy at this time can complicate early postoperative outcomes.

Mechanical irritation from chest-tube insertion can cause CABG patients severe pain, which can lead to shallow breathing, hypoventilation of the lungs, and suboptimal values of Po₂ and Pco₂.^2,3 In our study, there was no significant difference postoperatively between the groups in arterial blood gas analysis, respiratory rates, or oxygen saturations. These results indicate that either route for chest-tube placement can be used safely. This conclusion contrasts with the results of other studies, which showed that subxiphoid insertion caused less pain and led to significantly less impairment of pulmonary function after surgery than did intercostal in-sertion.^4,7 Hagl and colleagues⁴ reported that drainage of the left pleural space is equally effective through the subxiphoid or the intercostal approach. They also noted that the subxiphoid position is safer than the intercostal position because of the bleeding that can result from injury of the intercostal artery. There were no significant pulmonary sequelae, including hemothorax or pneumothorax, in either of our groups.

Two of our Group 1 patients had incomplete pleural drainage. The size, intrapleural location, patency, and suction efficiency of the tubes can all affect fluid drainage and the collection of residual effusion. In our Group 1 patients, 32F angled tubes were positioned with the tips in the costophrenic sinus. In our Group 2 patients, 32F straight chest tubes were inserted into the left hemithorax through the 6th intercostal space along the anterior axillary line with the tips toward the apex of the lung. All of the tubes were regularly cleared to avoid their obstruction by hematoma. The use of larger drains or stronger suction might have improved the pleural drainage in the 2 Group 1 patients. Of note, pleural effusions after CABG can be associated with decreased lymphatic drainage, inflammation of the pericardium, post-pericardiotomy syndrome, or trauma to the pleura, as well as to the chest tubes themselves.⁸ Any of these factors could have caused pleural fluid to collect after the chest tubes were removed. Accordingly, the residual effusion in our 2 patients might not have been related to the surgical approach for chest-tube placement.

Analgesic therapy after CABG can consist of acetaminophen, NSAIDs, tramadol, meperidine, or morphine. Although acetaminophen is comparatively the safest to use, it might not provide effective analgesia by itself.⁹ In our study, we routinely used NSAIDs, because they have been shown to decrease opioid-related nausea, vomiting, or sedation^6,10 and can be given before CABG to patients who do not have organ failure. We delivered additional doses of tramadol or meperidine, as necessary, to provide effective analgesia. We found no significant differences between the groups in the cumulative doses of necessary rescue analgesics. The low pain scores in both groups suggest that pulmonary morbidity decreases when analgesia is effective.

Study Limitations

Our study had several limitations. First, we studied a relatively small population that underwent CABG with sternotomy and harvesting of the left ITA. Second, a lack of spirometry precluded evaluation of postoperative pulmonary function beyond observing the apparently effective delivery of analgesia.¹¹ Third, the size and type of chest tubes might have affected the efficacy of pleural drainage, pain sensation, and associated morbidities. Fourth, the surgeon who recorded all outcomes was not blinded to the treatment groups. Finally, the timing of chest-tube removal might have affected the intensity of postoperative pain and the clinical outcomes.

In conclusion, we found that the subxiphoid and intercostal approaches were not significantly different with regard to early acute postoperative pain and pulmonary function after CABG.