Pulmonary and Biceps Function after Intercostal and Phrenic Nerve Transfer for Brachial Plexus Injuries

INTRODUCTION

Brachial plexus injuries with complete nerve root avulsions are irreparable and have no possibility for spontaneous recovery. Neurotization or nerve transfers are indicated for these injuries (Alnot, 1995; Chuang, 1995; Narakas and Hentz, 1988). In these procedures, a normal nerve with less important function is transferred to reinnervate a nerve which has an important function and has suffered irreparable damage. Various donor nerves have been described, including the spinal accessory nerve (Alnot, 1995; Chuang, 1995; Easton et al., 1983), the intercostal nerves (Alnot, 1995; Chuang, 1995; Chuang et al., 1992, 1993; Giddens et al., 1995; Krakauer and Wood, 1994; Nagano et al., 1989; Narakas and Hentz, 1988; Ruch et al., 1995; Waikakul et al., 1999), and the phrenic nerve (Chuang et al., 1993; Gu and Ma, 1996). Although intercostal and phrenic nerve transfer are both well established techniques for the treatment of severe brachial plexus injuries, the deterioration of pulmonary function after donor nerve division remains a major clinical concern. Krakauer and Wood (1994) suggested that there was a mild decline in pulmonary function in four of 12 patients after intercostal nerves transfer, although there were no subjective changes in respiratory status. Furthermore, Giddens et al. (1995) and Waikakul et al. (1999) reported that intercostal nerve transfer did not cause a significant reduction in respiratory function. Gu and Ma (1996) suggested that pulmonary function tests showed decreased pulmonary capacity for 1 year after phrenic nerve transfer but this then improved. Several studies have compared biceps recovery after different neurotizations (Chuang, 1995; Chuang et al., 1993, Hattori et al., 1997; Narakas and Hentz, 1988; Waikakul et al., 1999). However, there are no prospective comparisons of intercostal nerve and phrenic nerve transfer. In this study we compare the pulmonary and biceps function after intercostal and phrenic nerve transfer.

PATIENTS AND METHODS

The study is a pseudo-randomized comparative study. Patients aged 13 or more with a complete root avulsion type of brachial plexus injury which required neurotization were recruited consecutively from the Hand and Microsurgery Unit at our hospital. Patients were assigned to neurotization using either the third and fourth intercostal nerves or the phrenic nerve depending on the date of admission (odd versus even dates). All phrenic nerve transfers required a sural nerve graft whereas nerve grafts were never needed for the intercostal nerve transfers. All the operations were performed by the same surgeon. The time interval from the brachial plexus injury to surgical reconstruction was always less than 6 months. Patients were excluded if there was evidence of diaphragmatic paralysis, a history of chest trauma (such as haemothorax or rib fractures), associated injuries such as fractures or biceps muscle injuries to the same arm, or they were unable to return for follow-up visits or refused to participate in the trial. Between August 1998 and June 2000, 39 patients were recruited but three patients (two interscostal and one phrenic nerve transfer) did not return for follow-up visits and were thus excluded. Thirty-six (33 men and three women) completed the study. There were 19 patients in the intercostal nerves group and 17 in the phrenic nerve group.

Pre-operatively, all patients underwent physical examinations, chest X-rays for detection of diaphragmatic paralysis, electrodiagnostic studies and pulmonary function tests. Autospiropal spirometer was used to test the pulmonary function. The measurements recorded were forced vital capacity (FVC), forced expiratory volume in 1 second (FEV₁), vital capacity (VC) and tidal volume (TV). Each patient performed the pulmonary function tests on six occasions: pre-operatively and at 2 weeks and 3, 6, 9 and 12 months after surgery. We also studied the effect of body position on FVC, by comparing the FVC when standing and lying flat (supine). This was because posture can affect pulmonary function in phrenic nerve paralysis patients (Allen et al., 1985; Claugue and Hall, 1979; Gould et al., 1967).

Biceps function was also assessed at each follow-up visit, and the time of detection of the first biceps contraction, as well as biceps power, were recorded.

The primary outcomes in this study were the differences in pulmonary function before and after surgery, the time to first observation of a biceps contraction and biceps power during the first year after surgery. Continuous data were analysed with the Student’s t-test and analysis of variance. The discrete data were analysed with the chi-square test.

RESULTS

There were no statistically significant differences in the baseline characteristics of the intercostal nerves and phrenic nerve transfer groups (Table 1).

In the phrenic nerve group, the FVC, FEV₁, VC and TV were significantly reduced throughout the 1-year period of this study. In contrast, the FVC, FEV₁ and VC in the intercostal nerves group were only significantly reduced for 2 weeks after the operation (Tables 2–4). The TV in the intercostal nerves group was also only significantly reduced for 6 months after surgery (Table 5). Direct comparison of the FVC, FEV, VC and TV in the two groups showed that all were significantly lower in the phrenic nerve group than in the intercostal nerves group during the entire study period (Tables 2–5).

Body position had no effect on FVC in the intercostal nerve group, but significantly affected it in the phrenic nerve group (Figs 1 and 2).

Patients in the intercostal nerves group had significantly earlier (P=0.03) recovery of biceps contraction (mean, 195 day; range, 131 to 330) than those in the phrenic nerve group (mean, 262 days; range, 166 to 540). Motor recovery to grade 3 or greater in the biceps muscle was observed in 10 of the 19 patients in the intercostal nerves group and five of the 17 in the phrenic nerve group. Four patients in the phrenic nerve group had no recovery of biceps muscle at 1 year, but all the patients in the intercostals nerves group regained some motor function in this muscle (Fig 3) and, after our rehabilitation programme, could separate breathing from biceps function.

DISCUSSION

There are concerns that intercostal nerves and phrenic nerve transfer may reduce pulmonary function. However Giddens et al. (1995) found that intercostal nerve transfer did not result in significant reductions in FVC, FEV₁ and peak expiratory flow rate. Furthermore, Waikakul et al. (1999) reported no significant differences in the TV, VC and the FEV₁ to FEV ratio before and after intercostal nerves transfer. Our results are similar, except that we observed reductions in the FVC, FEV₁ and VC at 2 weeks after surgery. This may have been due to pain from the chest wound rather than harvesting two intercostal nerves. Gu et al. (1996) reported that the VC, total lung capacity, functional residual capacity and maximum ventilation volume were decreased for 1 year after phrenic nerve transfer, but then improved during the next year. Our findings are similar as we found that FVC, FEV₁, VC and TV were significantly reduced after phrenic nerve transfer throughout the period of the study. Fackler et al. (1967) reported that unilateral phrenic nerve section had remarkably little effect on respiratory function in patients with normal lungs. However, diaphragmatic paralysis may cause life threatening respiratory distress in infants and young children (Brouillette et al., 1986; Chuang, 1995; Haller et al., 1979) and unilateral phrenic nerve injury is associated with an abnormal pattern of respiratory muscle function during quiet breathing (Easton et al., 1983). Furthermore, inspiratory muscle strength is impaired in some of the patients, and this is worse when cardiopulmonary disease is present (Lisbosa et al., 1986). Therefore, phrenic nerve transfer should be avoided in infants, young children, and patients with cardiopulmonary disease.

In our study, we also found that body position had more effect on pulmonary function after phrenic nerve transfer than after intercostal nerves transfer. In the standing position gravity assists the downward movement of the diaphragm, whereas in the supine position the abdominal organs tend to push the diaphragm into the thorax, resulting in difficulty with breathing, especially deep inhalation (Allen et al., 1985; Claugue and Hall, 1979; Gould et al., 1967). Therefore, if the phrenic nerve is neurotized, we recommend that patients should rest in Fowler’s (semi-sitting) position postoperatively.

Previous studies of intercostal nerves transfer have reported good or excellent recovery of biceps muscle function (motor power grade 3 or more) in 47% to 67% of cases (Chuang et al., 1992; Ruch et al., 1995). After phrenic nerve transfer, Gu and Ma (1996) reported biceps motor function of grade 3 or more in 85% at a follow-up of more than 2 years. In our study ten of 19 patients regained biceps muscle power of grade 3 or more after intercostal nerves transfer, compared with five of the 17 patients in the phrenic nerve group. In addition the functional recovery of biceps muscle in the intercostal nerves group was earlier than that in the phrenic nerve group. This might have been due to the shorter distance of nerve regrowth required after intercostal nerve transfers, and also because all the phrenic nerve transfers required a sural nerve graft.

In conclusion, although none of the patients in either group were troubled with respiratory symptoms during normal activities of daily living, we recommend the use of intercostal nerves, rather than the phrenic nerve, for nerve transfer in brachial plexus injuries. We particularly feel that the use of the phrenic nerve should be avoided in children, or patients with pulmonary diseases.