Brachial plexus block, a kind of local anesthetic, is effective in shoulder and upper limb surgery with or without general anesthesia [19]. It is likely that the use of a brachial plexus block before surgery could not only reduce postoperative opioid consumption, but also effectively promote rapid postoperative recovery and reduce the hospital length of stay [20]. Supraclavicular plexus block or an interscalene plexus block are the most frequently used nerve blocks, although diaphragm paralysis is a known complication of both blocks [18]. Since the interscalene plexus is very close to the phrenic nerve at a high level of the pouch, the risk of diaphragm paralysis is higher [21]. Also correlation between the occurrence of diaphragm paralysis and supraclavicular plexus block has been observed [22]. This is mainly related to the contralesional diffusion of local anesthetics and applying pressure on the proximal nerve trunk does not contribute to reducing the paralysis rate. The axillary plexus is far enough away from the edges and far enough away from the phrenic nerve; however, it is not always a reliable method on the elbow or above, and it may miss the radial forearm and musculocutaneous nerve. Respiratory function during the early postoperative course, comparing supraclavicular plexus block and interscalene plexus block has rarely been studied or investigated.
The brachial plexus is close to the abdominal branch of the C5 nerve at the cricoid cartilage, with an average distance of 1.8 mm. On average, the distance between the brachial plexus and phrenic nerve increases by 3mm for every 1cm of the neck to the caudal end. Thus, it is easy to infer that the phrenic nerve is highly vulnerable to also be blocked during an interscalene plexus block. Diaphragm mobility is first affected, then the blockage involves the lung function. Our preliminary study showed that the diaphragm mobility of group I was lower than that of group S at T30min, T4, T8, and T12, which may be related to the position of the phrenic nerve. At the same time, the diaphragm mobility of group S was not statistically different between the level before the block (T0) and T12 by the method of one-way ANOVA analysis, with Dunnett’s post-hoc test, but the difference was significant in group I. In terms of the recovery speed of diaphragm mobility, the results implied that the changes of diaphragm mobility in group S returned to a near-normal level after a 12-hour recovery time; this phenomenon, however, was absent in group I, which provided a reference for blinding postoperative observation period. This was probably related to the recovery time of the sensory block, but, as a whole, the recovery of diaphragm mobility was delayed. Many inherent causes may have contributed: uncontrollable mechanical damage to the phrenic nerve, the toxicity of anesthetics, the vasoconstrictor effect of ropivacaine, and so on. Ultrasound technology is very important to demonstrate diaphragmatic dysfunction, not only because we could rapidly evaluate patients but also because it could avoid unnecessary radiation exposure [23]. We used ultrasound technology to quantitatively measure the movement of the diaphragm before and after brachial plexus block and analyzed the incidence of phrenic nerve block. Due to the obstruction of the spleen and gastrointestinal cavity organs on the left side of the diaphragm, the diaphragm had an unclear outline, making it difficult to measure the movement of the left side of the diaphragm [24].. It is easier to observe the movement of the diaphragm through the right side. In order to approximate the actual clinical situation, patients with right brachial plexus block were not specifically selected.
The main observation indicators of our experiment indicated that the incidence of the phrenic nerve block of group I was 67%, and the rate of diaphragmatic palsy of group S was 41%. Incidence of HDP after ISB from our study was lower than previous studies [22, 25]. This may be due to different kind, concentrations, and doses of local anesthetics. Although our study had a small study sample, this is consistent with other studies This may not be sufficient to develop respiratory distress with accessory respiratory muscles assisting [26]. Besides, a previous study has suggested that the movement of the contralateral diaphragm can be compensated for synchronously with the change in the movement of the blocked diaphragm according to contralateral diaphragm assessment [27]. However, due to the delayed diaphragmatic palsy, the block may have serious consequences if it was implemented in patients with respiratory diseases or obesity [28, 29]. When not supported by adequate supervision, patients with respiratory depression may be ignored.
Decreased diaphragm activity affects the patient’s breathing. The invasive operation of blood gas analysis of patients during a short operation time is not in line with the principle of benefiting them. Therefore, we evaluated the patient’s respiratory function by non-invasive lung function measurements. General anesthesia was applied to these patients after the nerve block, which might increase patient comfort. Since the pulse oximetry were affected by general anesthesia, we do not include it in the detection index. Other studies have shown that as long as the patients were in good physical condition, unilateral phrenic nerve block was less likely to cause a severe drop in pulse oximetry [13, 22].
Pulmonary complications mostly occur in the postoperative period. Our results showed that pulmonary function testing (FVC and FEV1) of both groups was impaired to different extents after the blocks. Diaphragmatic paralysis could not increase the transverse and anteroposterior diameter of the thoracic cavity, so that the volume of the lungs could not be increased correspondingly [30]. As a result, the patient's lung volume at the end of inhalation was reduced, the initial length of the muscle fibers was shortened, and the ability of muscle contraction was reduced, leading to a decline in lung function. In the one-way repeated-measures ANOVA analysis, FVC and FEV1 of group S was recovered after 12 h of blockade in our study, while in group I they were not, which seemed to go hand in hand with diaphragm mobility. The change of lung function may be related to the degree of diaphragm paralysis [31, 32]. Although possibly related to regional anesthesia, ventilation under general anesthesia could also be associated with exposure to lung injury. However, this should have little effect on the results, since it should be equally distributed between group S and group I.
The differences in analgesic effect just failed to reach statistical significance at T30min, T4, T8, and T12. The recovery of motor and sensorimotor functions was not different between the two groups. We found no significant difference in the adverse events between the two groups, even though Horner syndrome, nausea, and vomiting had a trend toward a difference between the two groups. We hypothesized that the incidence of nausea and vomiting may be affected by the anesthesia induction was performed by endotracheal intubation which can bias these results.
Our study has several limitations. Considering compliance to the treatment of the patients, our measurements pertained only to the 12 h postoperative period, and thus the recovery time of respiratory function in group I needs to be explored in larger, longer-term studies.