can be generalised. Patients with secondary nose bleeding
(traumatic, cancer or related to sniffing narcotics) were
excluded, because their healing process was assumed to be
determined by the cause of the bleeding. A new trial would
be required to establish whether patients with secondary nose
bleeding may be mobilised without affecting their risk of new
bleeding episodes. The same applies to patients who had
been admitted with primary nose bleeding within three
months prior to the trial period. These cases were excluded,
because the initial treatment may have resulted in pressure
necrosis of the mucosa and cartilage and thus represent
particularly vulnerable cases (Delank 2006, Viehweg et al.
2006).
The age of the participants and the number of new bleeding
episodes tallied with the information we had previously
recorded in the ward (unpublished observation). We therefore
assume that the participating group is a representative
sample of all epistaxis patients admitted to the ward.
Furthermore, as the ward serves the entire region and receives
patients with less as well as more complicated cases, we
consider the trial participants as representative of patients
admitted to hospital with epistaxis in Denmark.
The two groups were different with respect to the number
of participants with diabetes and the number of women. In
the study group, two of five diabetic participants experienced
new bleeding, while the same was true in six of 12
participants in the control group. Overall, 10 of the 21
women in the study group experienced bleeding, compared to
seven of 15 women in the control group. These differences in
the number of women and participants with diabetes therefore
had no bearing on the primary outcomes.
Similarly, additional analysis for confounding factors such
as arterial hypertension, bleeding disorders, anticoagulant
therapy and diabetes showed no differences with respect to
renewed bleeding patterns; the material was inadequate for
stratification. Only cauterisation of nasal mucosa had a
significant effect on the number of new bleeding episodes,
which was to be expected, as this treatment directly affected
the bleeding spot. Additional studies would be needed to
elucidate in more detail, whether specific groups of patients
need differentiated treatment regimes.
The number of complications was larger for the control
group than for the study group (Table 4), but the overall
numbers were small; our results cannot be taken as a
demonstration that bed rest resulted in more complications
compared with mobilisation, although this could be argued
theoretically. On the other hand, it may be assumed that the
general health of mobilised participants was improved and
that they sustained a lesser risk of being affected by
complications when compared to bed rest patients.