PATIENTS AND METHODS

Patients

All patients treated in our stroke unit within a 10 month period who needed tube feeding were included in the study. The study includes patients already published in a preliminary report.¹³ Swallowing was assessed with the 50 ml water swallowing test.¹⁴ Coughing during or for 1 min after completion of the test or the presence of a post-swallow wet, hoarse voice quality were scored as abnormal and indicated a need for temporary tube feeding. In all patients the severity of the stroke was measured with the National Institute of Health Stroke Scale (NIH scale). The study was approved by the local ethics committee, which waived the need for informed consent.

Reflex placement of NGTs

A thin catheter was inserted through the nostril with its tip being placed in the oropharynx. The NGT was placed through the other nostril in approximately the same position. The swallowing reflex was induced by bolus injection of 0.5–2.0 ml of distilled water through the thin catheter.¹⁵ At the onset of swallowing, which was identified by observation of the characteristic upward laryngeal movement, the NGT was moved forward.

Study design

For evaluating the efficiency of the new method, patients in whom a NGT had to be placed were divided into two groups. The control group comprised those patients in whom the conventional approach was successful. Patients in whom the conventional approach failed were allocated to the study group. At least four attempts and the use of known facilitating measures such as increasing the rigidity of the tube by refrigeration and tilting the patient’s head forward were required before the application of the new method was allowed.¹⁶

For evaluating the tolerability of the new method, changes in heart rate, systolic arterial blood pressure and oxygen saturation in response to either method of tube placement were scored in a further group of patients needing tube feeding because of dysphagia. For this investigation, we took into account only patients for whom, because of self-extubation or tube obstruction, at least two NGT placements within 48 h were necessary, and therefore both methods could be studied in close succession. In half the patients, the conventional method was performed first, and in the other half the new method was used. The number of attempts until achievement of correct placement was scored. The presence of concomitant medication modulating blood pressure or heart rate was noted.

Statistics

Unpaired data was compared with the Mann-Whitney U test, paired data was analysed with the Wilcoxon matched pair test. The level of significance was set at p<0.05.

RESULTS

During the study period, 386 patients were treated in our stroke unit. Sixteen (15.4%) of 104 patients who needed tube feeding because of disturbed swallowing were selected for the study group. The mean age and severity of stroke did not differ significantly between the two groups (table 1). Clinically, ten patients in the study group had severe aphasia and buccofacial apraxia because of infarction of the left middle cerebral artery (MCA) or left sided intracerebral haemorrhage (ICH). Eight patients, two with infarction of the left MCA, three with infarction of the right MCA, and three with brainstem infarction, had a markedly reduced level of consciousness. In these patients, the conventional approach repeatedly resulted in tracheal positioning of the tube or its coiling in the mouth. By using the swallowing reflex, the tube was successfully placed in 14 of 16 patients (87.5%). The two failures occurred in patients with a left MCA infarction and a brainstem infarction respectively, in whom the swallowing reflex could not be provoked.

Twelve patients (mean age 69.0 (5.9); mean NIH 15.9 (4.7)) were included in the second part of the study. Five had suffered an infarction of the left, and four of the right MCA. One patient each suffered an infarction of left MCA and ACA, a brainstem infarction, or a right sided ICH. During both placements, a continuous infusion of urapidil was given to two patients and verapimil to one. As stated above, the conventional method was used first in six patients, while in the other six patients the NGT was initially placed using the new method.

The new method enabled tube placement in nine patients in the first and in three patients in the second attempt, making an average of 1.3 trials, whereas the conventional approach was successful only after an average of 2.2 trials (p<0.01; table 2). This difference was mainly due to repetitive gagging and insufficient voluntary swallowing of the NGT during the conventional approach. The baseline parameters were similar for both groups. In both the conventional and the new method, placement resulted in a significant increase in heart rate and systolic arterial blood pressure, whereas oxygen saturation remained stable. However, these haemodynamic changes were significantly larger for the conventional than for the new method (average maximum systolic arterial blood pressure of 176 mmHg v 154 mmHg; average maximum heart rate of 108 beats/min v 94 beats/min, respectively) (table 2).

DISCUSSION

Our study confirms that the placing of NGT by inducing the swallowing reflex is a useful alternative if the conventional method fails. In 14 of 16 patients (87.5%) in whom the conventional approach repeatedly resulted in tracheal positioning of the tube or its coiling in the mouth, the new method was successful. Equally encouraging results using the reflex placement were achieved by Inoue et al¹⁷, who reported a success rate of 82.6% in a more elderly collective of stroke patients (mean age 86.7 years).

A comparative study of the cardiovascular responses to both methods showed that the reflex placement was less distressing for the patients, leading to a significantly smaller increase in heart rate and systolic blood pressure. The main reason for this seems to be the fact that with this method significantly fewer attempts were needed (1.3 v 2.2) until correct placement of the tube was achieved.

The main drawback of our procedure is that it may be of only limited use in patients with brainstem infarction as they are likely to lack a sufficient swallowing reflex.

Apart from the argument of increasing the patients’ comfort, the avoidance of intermittent peaks of intracranial pressure induced for example by coughing or gagging constitutes a basic principle of care in patients with space occupying brain lesions.¹⁸ Therefore, even if the new method requires a larger effort by necessitating the co-operation of two medical staff members, we recommend its use in patients who have suffered a stroke and need tube feeding due to dysphagia.