In response to Drs. Bernstein and Scott and colleagues, only the lowest dose of vecuronium required to prevent patient ventilator dyssynchrony was used. Neuromuscular blockade was discontinued at least twice daily to assess neurologic function and the continued need for paralysis. The dosage of vecuronium required to support long-term mechanical ventilation must be individually titrated. Patients with hepatic or renal insufficiency may require lower doses because of accumulation both of vecuronium and its 3-desacetylvecuronium metabolite. Such patients were excluded from our study. Tolerance, however, has been reported with prolonged use [1] with doses of vecuronium as high as 32 mg/h. These levels are required to maintain adequate relaxation for controlled ventilation.

As pointed out by Drs. Barnette and Carlsson, clinically significant blockade of nerve transmission begins when 75% of receptors are occupied by neuromuscular blocking agents. This degree of paralysis is often sufficient to prevent patient-ventilator dyssynchrony and also lowers airway pressures [2]. Complete diaphragmatic paralysis is rarely required; if it is required, the effects of neuromuscular blockade may be easily monitored with train-of-four stimulation of the orbicularis oculi muscles, which have a sensitivity to neuromuscular blockade similar to the diaphragm [3].

As to the development of myopathy and its relation to steroids, eight of the ten patients received intravenous methylprednisolone; the mean dose was 3880 mg. Steroid use did not differ between the patients who developed neuropathy and those who did not (P = 0.17). Patient 6 developed a polyneuropathy but did not receive any steroids. None of the patients in our study showed elevated levels of creatine kinase, whereas the patient cited by Scott and colleagues [4] had a fourfold elevation of creatine kinase levels as did a majority (10 of 16) of patients in the review by Hirano and colleagues [5]. All five patients with polyneuropathy in our study showed reduced maximum motor conduction velocity in multiple nerves. Additional electromyographic signs were compatible with the decreased motor nerve conduction velocity, including reduced evoked compound muscle action potentials, changes in muscle activities obtained by needle electrode, and decreased maximum sensory nerve conduction velocity with reduced compound sensory nerve action potentials. A previous report of polyneuropathy in patients who had prolonged neuromuscular blockade has shown neuropathic changes in muscle and nerve biopsies [4]. The clinical, blood chemistry, and electromyographic data are compatible with a polyneuropathy and not a myopathy.