Five years of Telemedicine in Cystic Fibrosis Disease

Clin Ter 2009; 160 (6):457-60

Original article

Five years of Telemedicine in Cystic Fibrosis Disease

S. Bella1, F. Murgia1, A.E. Tozzi2, C. Cotognini3, V. Lucidi1
1

Cystic Fibrosis diagnosis and treatment Unit, 2Epidemiology Unit, 3Information Systems and Organization Department, Childrens
Hospital Bambino Ges, Research and Care Institute, Rome, Italy

Abstract

The natural history of Cystic Fibrosis (CF) is characterized by a progressive lung destruction, caused by obstruction
Aims. We studied the effect of Telehomecare (THC) in a group of
of the airways due to dehydrated thickened secretions, reCystic Fibrosis (CF) patients, with the aim to early recognize the resultant endobronchial infection and an exaggerated inflamlapses of pulmonary infections. Data obtained with Vivisol (OXYTEL)
matory response leading to development of bronchiectasis
instrumentation were collected from 2001 to 2005.
and progressive obstructive airways disease (4). Prevention
Materials and Methods. The study has involved 17 patients (11
and control of lung infections is one of the main objectives
f, 6 m) affected by CF, treated with THC, in addition to the usual
of therapy in CF patients with the aim to slow down the
therapeutic protocol, for an average period of 29.6 months 13.5.
progressive decline in pulmonary function with time (5).
The mean age for THC enrolment was 15.74 years 5.8. As controls,
An acute decrease of FEV1 >10% is considered a signithe study has involved the same patients during the 12 months prior
ficant sign of infectious pulmonary relapse (6).
to THC start-up and 28 patients affected by CF treated at our Unit (13
Many researchers demonstrated that, in case of infectious
f, 15 m; average age 14.77 5.22).
relapse, pulmonary function modifications often precede
Results. The results show a statistically significant decrease of
the clinical symptoms and that monitoring variations in
outpatient accesses and increase of therapy cycles, and a trend of
pulmonary function can be useful in children and in adults
higher stability of the respiratory function, in THC treated subjects
(7, 8).
compared to controls.
Early recognition of infectious relapse allows to timely
Conclusions. Our study suggests that THC programs may not be
administer an antibiotic therapy, to prevent serious complioptimally accepted by CF patients, and that its use seems to increase
cations, and to use less aggressive therapies (9).
in general the rate of access to health care without demonstrating any
Purpose of this study was to assess the effect of THC in
clear effect of pulmonary function.
the follow-up of CF patients, by systematically monitoring
respiratory parameters (O2 saturation during the night and
cystic fibrosis, telemedicine, telehomecare
Key words:
spirometry), to early detect pulmonary infectious relapses,
and to measure the impact on respiratory function over
Introduction
time.
Innovative technologies and informatics applied to medicine offer both health operators and patients a wide range of
services that have changed the traditional concept of health
care. In recent years, the availability of handy equipment,
easy to transport and use, and suitable to collect and transmit
various clinical data, has resulted in a fast development of
Homecare. The earliest application of Telemonitoring involved the follow-up of acute patients affected by arrhythmia
or heart failure, diabetes, acute respiratory insufficiency as
bronchial asthma, the control of breast-feeding mothers during lactation, the assessment of post surgery patients (1).
Only recently Telehomecare (THC) became an opportunity for the follow-up of chronic diseases such as cardiopulmonary diseases, bronchial asthma and heart failure
(2), although the impact of Telemonitoring on patients
conditions still remains uncertain (3).

Materials and Methods

Study design

We performed an open label trial in a population of CF

patients followed in a reference centre for CF from 2001 to
2005. Patients were considered eligible to enter the study
if presented multiple infectious lung relapses (more than
3-4 episodes in a year) and/or significant decrease of mean
FEV1 (more than 10% in a year).
The intervention study consisted in administering THC
in adjunct to standard therapy. THC was assigned to the
first patient seen in the week, who satisfied the eligibility
criteria. A group of controls was chosen among patients
seen on the same week, matching for respiratory function,

Correspondence: Dr. Sergio Bella, Centro di diagnosi e cura per la Fibrosi Cistica. Ospedale Pediatrico Bambino Ges, Piazza S. Onofrio, 4,
00165 Rome, Italy. E-mail: bella@opbg.net

S. Bella et al.

458
bacterial colonization, sex, age, and complications. The main
outcome measure we considered in the study was FEV1 values over time. To standardize for individual characteristics,
FEV1 Zscores were calculated considering average FEV1
and standard deviation measured in the 12 months before
entering the study. FEV1 was then measured during follow
up, and Zscores calculated and compared in the two groups.
All patients, regardless of intervention assignment, were
provided antibiotic treatment and the therapies according
to current treatment guidelines (11).
Telemonitoring intervention

number of events per 100 person months. Differences in

rates have been studied comparing rate ratios and their 95%
confidence intervals with the mid-p exact test. Continuous
variables, including age at enrolment, FEV1 at enrolment,
and follow up duration were compared by the Student T test,
once verified the normality of the distribution.
To analyse FEV1 variations we considered monthly average Zscores by THC use. Observations were interpolated
through simple linear regression to look for trends. To adjust
for multiple measures in the same patient, we compared the
two groups by robust regression.

Telemonitoring was performed using Vivisol OXYTEL Results

M32 (www.vivisol.com), a digital multi-parametric recorder
with integrated saturimeter. OXYTEL M32 is a multi-channel
Out of 60 CF patients enrolled in the study 30 were
recorder able to receive data from external devices, such as assigned to THC and 30 to the routine care program only.
spirometers. Spirometries were obtained using a Vivisol One Thirteen out of the 30 patients assigned to THC have been
Flow Spirometer linked with OXYTEL by cable. Both pasubsequently excluded: 11 because of a short period of trantients and parents were trained by the centre physicians, part smission (< 6 months) or because they provided incomplete
of the multidisciplinary CF care team, using the equipment. registration, whereas 2 patients had unpredictable complicaOXYTEL can record and store overnight SaO2, and pulse
tions and have died during the study, with a resulting dropout
rate. Home recording of SaO 2 was carried out through all
of 43.3%. Two control patients withdraw participation in
night. Spirometry was performed in the morning, after chest the study within six months. We analysed thus the data of
physiotherapy for mucous drainage. Data were automatical- 17 CF subjects who received THC for a minimal period of
ly sent by modem via public telephone line to a dedicated
7,04 months and of 28 control patients.
Personal Computer located in the CF Unit, and decrypted
The general characteristics of patients participating in
by a dedicated software. Data were read and interpreted
the study are illustrated in Table 1. Although assignment to
by CF Unit medical team during work shifts. Patients were THC was not randomized, the baseline characteristics of the
reached by a physician by telephone and the clinical situation two groups were similar and none of the comparisons were
evaluated. Depending upon conditions, further data transmis- statistically significant.
sions could be requested or oral antibiotic therapy straight
Table 2 shows a comparison of mean hospitalization
prescribed. Otherwise patients could be recalled in hospital rates, mean outpatient rates, and number of therapy cycles
for more surveys and therapies. Data collection of respiratory in the 12 months before enrolment in the trial and during
parameters through Telemonitoring was performed generally the follow up.
twice a week. Patients could autonomously decide to transmit
To assess the compliance of patients assigned to THC,
depending on the individual clinical condition.
during the follow-up period we considered the number of
We used data from OXYTEL instrumentation only
days in which patients transmitted overnight SaO2 value.
for monitoring variations in pulmonary function at home,
The average compliance (transmissions/patient/day) was
with the aim to early recognize the relapses of pulmonary
52.4%.
infections.
THC patients had a higher outpatient visit ratep=0.024)
(
Acute FEV1 variations less than 10% from the individual and a higher number of therapy cycles ( p=0.01) compared
baseline were considered as an indication for antibiotic
with controls during the follow-up.
treatment for the intervention and control groups.
Figure 1 shows the FEV1 monthly means in THC patients
and controls during the follow- up, expressed as Zscore.
Statistical analysis
The trend over time is showed both in THC subjects and in
controls by a regression line.
Beside FEV1 we calculated also outpatient visit rate and
Both groups showed a progressive decrease of FEV1 over
hospitalization rate during the follow up together with rates time and no statistically significant difference was detected
of intravenous antibiotic therapy and we expressed them as when comparing the two groups over time. However, the

Table 1. General characteristics of subjects exposed and not exposed to THC.

Number of patients analysed

Males/Females
Age at enrolment. years (mean SD)
FEV1 at enrolment % of expected value (mean SD)
Follow-up duration. months (mean SD)

Telehomecare

Controls

17
6/11
15.74 5.8
67.48 21.28
29.30 13.32

28
13/15
14.77 5.22
76.47 24.56
30.13 20.49

0.463
0.338
0.218
0.882

Five years of Telemedicine in Cystic Fibrosis Disease

459

Table 2. Hospital access and therapy cycle rates before and after the intervention.

Baseline

THC
Mean hospitalization rate
(Admissions per 100 person
months)

Controls

13.73

15.2

55.39

44.61

21.08

15.2

Mean outpatient visit rate

(Visits per 100 erson months)
Mean therapy cycles
(Cycles per 100 person month)

Rate ratio;
95% CI;
P
0.90;
054-1.51;
0.699
1.24;
0.94-1.64;
0.124
1.41;
0.87-2.21;
0.165

THC

During intervention
Rate ratio;
95% CI;
Controls
P

26.5

24.09

46.38

37.14

22.29

15.26

1.1;
0.86-1.41;
0.451
1.25;
1.03-1.52;
0.024
1.46;
1.09-1.96;
0.01

by additional interventions. Certainly compliance figures

must be considered when planning such an intervention and
possible related clinical trials.
Subjects monitored through THC had a higher number of
therapy cycles compared to controls. On the other hand, no
statistical difference was detected in mean FEV1 values over
time when comparing the two groups, although a tendency
toward lower FEV1 values was detected in controls.
The decrease in the number of outpatient accesses
(p=0.02) and the increase in the number of therapy cycles
(p=0.01) between treated subjects and controls are statistically significant in both cases.
Therefore, statistical tests are significant only in the
intervention period, because the period of observation was
Fig. 1. FEV1 monthly averages and trend lines.
longer than baseline period and confidence limits of rate
are narrowed. Despite these rates seem significant only during intervention, they are similar in the two periods. This
means that was no effect of THC treatment on the measures
considered.
We observed an increase in the number of hospitalizalinear trend calculated on Zscore FEV1 values of controls
tions from the pre-intervention period to the follow up period
tended to lower values with follow-up time.
in all patients. This was likely the effect of the change of
Moreover, several mean FEV1 values in patients without follow up protocols for CF patients care, during the period
THC were lower than 2 standard
of the study.
deviations with respect to the 12 months before enOur results should be interpreted with caution due to the
rolment, whereas the minimum mean Zscore FEV1 value
lack of randomisation and a limited number of CF subjects
recorded in THC patients was -1.46.
participating in the study. Furthermore, it is possible that
THC instrumentation was unintentionally given to more unstable subjects, with the aim to improve the follow-up control
Discussion
over them. This may have likely produced a bias toward the
lack of differences in FEV1 trends in the two groups.
Our study suggests that THC may be helpful in reduHowever, in the THC-treated group we noted a higher
cing the decline of respiratory function in patients who are
stability of FEV1 values according to lower SD values of
compliant with THC. Our data are consistent with previous FEV1. This trend, if confirmed, might suggest a situation of
studies of Finkelstein (12), showing the non-interference of higher stability of the FEV1 and, therefore, of the respiratory
home monitoring and daily diary recording with physical or function, in THC treated subjects.
psychological status of FC patients, and of Magrabi (13),
In conclusion our study suggests that THC programs
showing the feasibility of THC for monitoring CF.
may not be optimally accepted by CF patients, and that its
Compliance to THC was an issue in our study. Only
use seems to increase in general the rate of access to health
56,6% of initially enrolled subjects were able to continue
care without demonstrating any clear effect of pulmonary
THC for a period >6 months. During the follow up period, function.
we then recorded an average compliance (transmissions/paFurther larger randomized studies on the effect of THC
tient/day) of 52,4%. We may speculate that chronic patients on pulmonary function are needed to clarify if it represent
who receive several different therapies, may be overwhelmed a real benefit for CF patients.

460

S. Bella et al.

References
1. Scalvini S, Vitacca M, Paletta L, et al. Telemedicine: a new
frontier for effective healthcare services. Monaldi Arch Chest
Dis 2004; 61:226-33
2. Meystre S. The current state of telemonitoring: a comment
on the literature. Telemed J E Health 2005; 11:63-9
3. Par G, Jaana M, Sicotte C. Systematic review of Home
Telemonitoring for Chronic Diseases: The Evidence Base. J
Am Med Inform Assoc 2007; 14(3):26977
4. Flume PA, OSullivan BP, Robinson KA, et al. Cystic Fibrosis
Pulmonary Guidelines: chronic medication for maintenance
of lung health. Am J Respir Crit Care Med 2007: 176:95769
5. Que C, Cullinan P, Geddes D. Improving rate of decline of
FEV1 in young adults with cystic Fibrosis. Thorax 2006;
61:155-7
6. Ramsey BW, Farrell P M, Pencharz P. Nutritional assessment
and management in cystic fibrosis: a consensus report. The
Consensus Committee. Am J Clin Nutr 1992; 55(1):108-16
7. Davis S, Jones M, Kisling J, et al. Comparison of normal

8.

9.
10.

11.

12.

13.

infants and infants with cystic fibrosis using forced expiratory flows breathing air and heliox. Pediatr Pulmonol 2001;
31:17-23
Mohon RT, Wagener JS, Abman SH, et al. Relationship of
genotype to early pulmonary function in infants with cystic
fibrosis identified through neonatal screening. J Pediatr 1993;
122:5505
Rajan S, Saiman L. Pulmonary infections in patients with
cystic fibrosis. Semin Respir Infect 2002; 17:47-56
Urquhart DS, Montgomery H, Jaff A. Assessment of hypoxia
in children with cystic fibrosis. Arch Dis Child 2005; 90:
1138-43
Clinical Practice Guidelines for Cystic Fibrosis Committee.
Clinical practice guidelines for cystic fibrosis. Bethesda, MD:
Cystic Fibrosis Foundation 1997
Finkelstein SM, Wielinski CL, Kujawa SJ, et al. The impact
of home monitoring and daily diary recording on patient
status in Cystic Fybrosis. Pediatr Pulmonol 1992; 12:3-10
Magrabi F, Lovell N. H, Henry RL, et al. Designing home
telecare: a case study in monitoring cystic fibrosis. Telemed
J E Health 2005; 11:707-19