ORIGINAL PAPER
The use of non-invasive ventilation in very old patients with hypercapnic acute respiratory failure because of COPD exacerbation A. Nicolini,1 M. Santo,2 L. Ferrera,3 M. Ferrari-Bravo,4 C. Barlascini,5 A. Perazzo1
1
SUMMARY
What’s known
Aims: We prospectively enrolled 207 patients (121 were 75 or older and 86 younger than 75) who were admitted to three Respiratory Monitoring Units. The primary outcomes were intubation and mortality rates; the secondary outcomes were changes in arterial blood gases analysis, non-invasive ventilation (NIV) duration and length of hospital stay. Results: Hospital mortality was similar in the two groups, as were intubation rates. The proportion who died in the very old patient group was 19.8% (24/121) vs. 10.4% (9/86) in the younger group. Intubation rate was 10.7% (13/121) in the very old patient group and 11.6% (10/86) in the younger group. The presence of comorbidities, the severity of illness (SAPS II), the level of consciousness, NIV failure (intubation), absolute value of pH prior to NIV, as well as the changes in pH and paCO2 and PaO2/FiO2 after 2 h of NIV, were the variables associated with higher mortality. Very old patients had significantly higher NIV duration than younger patients (69.0 47.0 vs. 57.0 27.0 h) (p ≤ 0.03) and hospital stays (11.6 3.8 vs. 8.4 1.4) (p ≤ 0.02). Conclusions: The use of NIV in very old patients was effective in many cases. Endotracheal intubation after NIV failure was not efficacious in either group.
Introduction The use of non-invasive ventilation (NIV) as alternative to endotracheal intubation in patients with chronic obstructive pulmonary disease (COPD) exacerbations or acute cardiogenic pulmonary oedema is supported by randomised controlled trials and metaanalyses (1). More recently, NIV has been employed successfully in settings outside the intensive care unit (ICU), including hospital wards (1,2). In these settings, NIV showed a success rate for COPD patients similar to that reported in ICU (1,3,4). The choice of NIV aims to avoid complications, particularly in patients with complicated pathological situations (5). Patients aged 75 years or older, also referred to as ‘very old patients,’ are potentially ‘good candidates’ for less invasive management (6). The proportion of elderly persons among hospitalised patients, including ICU admissions, is rapidly increasing in developed countries. Very old patients represent 10–15% of ICU admissions (6–8). The management of critical respiratory illness in old patients is of increasing importance.
Non-invasive ventilation (NIV) is an alternative choice to intubation in hypercapnic respiratory failure due to exacerbations of COPD. It avoids complications associated with intubation. Very old patients are good candidates for NIV.
What’s new This prospective study has demonstrated that NIV in very old patients achieved a high rate of success. Endotracheal intubation after NIV failure is of questionable value in very old COPD patients. NIV success is inversely related to the number and the severity of co-morbidities, level of consciousness and directly related to an early resolution of the respiratory distress.
Furthermore, NIV is utilised for the respiratory support of patients with a ‘do not intubate’ advance directive (9–12). Certainly, ‘do not intubate’ cannot be considered as an indication for NIV; however, the use of NIV as a palliative measure is increasing (6). The results of short-term use are promising, but specific data for long-term outcomes and mortality are limited (6,13–15). These considerations prompted us to study the value of NIV in very old patients (aged ≥ 75 years) presenting with acute respiratory failure (ARF). We collected data in the hospital and 6 months after discharge. A younger group (< 75 years of age) was used for comparison.
Respiratory Medicine Unit, ASL4 Chiavarese, Sestri Levante, Italy 2 Emergency Medicine Department, ASL4 Chiavarese, Lavagna, Italy 3 Respiratory Diseases Department, Villa Scassi Hospital, Genova, Italy 4 Hygiene and Public Health Department, ASL4 Chiavarese, Chiavari, Italy 5 Forensic Medicine, ASL4 Chiavarese, Sestri Levante, Italy Correspondence to: Nicolini Antonello, Respiratory Medicine Unit, Via Terzi 43 – General Hospital, 16039 Sestri Levante, Italy Tel.: + 00390185329145 Fax: + 00390185329935 Email: antonello.nicolini@ fastwebnet.it
Disclosure None.
Methods The study was conducted in the Respiratory Monitoring Unit (RMU) and Emergency Medicine Unit of ASL4 Chiavarese, Italy and the RMU of the Respiratory Diseases Department of Villa Scassi Hospital, ASL3 Genovese, Italy, from May 2011 to June 2013.
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529. doi: 10.1111/ijcp.12484
1523
1524
NIV in elderly people
The study was approved by the Ethics Committee of ASL4 Chiavarese (no. 502 – 11/9/2010). These three units each have four non-invasive monitored beds and admit patients with severe respiratory failure who need NIV. We prospectively enrolled 207 patients with exacerbations of COPD, severe hypercapnic respiratory failure (ARF) and moderate to severe acidosis. One hundred and twenty-one of the 207 patients were 75 or older; 86 were younger than 75 years. All the patients were treated with non-invasive ventilation. One hundred and ninety-five patients gave written informed consent. In the other 12 patients who were unable to give written informed consent, the closest family member gave written informed consent. Diagnosis of COPD was based on the presence of airflow obstruction observed in previous pulmonary function tests. The severity of disease was based upon GOLD criteria (16). The RMUs had ventilators specifically designed for NIV (Philips Respironics Vision BiPap, Philips Respironics V60) and invasive ventilation (Philips Respironics V 200 Esprit, Versamed Ivent 201). Thus, it was possible to switch from NIV to invasive ventilation at any moment. Criteria for starting NIV as well as criteria of exclusion were previously described (17–19). Furthermore, patients enrolled for NIV did not present any criteria for emergency intubation (e.g. respiratory pauses, agitation requiring sedation, hemodynamic instability, systolic blood pressure < 90 mmHg or heart rate less than 40 beats/min). Pressure support ventilation was the preferred mode delivered by oro-nasal (full-face) or total face mask. We collected demographical and daily life data (age, sex, dyspnoea according to the classification of the MMRC, number of drugs taken per day, daily life limitations measure according to the Barthel test) and previous medical history (number of comorbidities, Charlson index, oxygen therapy, home nocturnal or diurnal ventilation or both, previous admissions to RMU or Intensive Care, and lung function. Moreover, the reasons for admission to our units and the severity of the respiratory signs were recorded. These included respiratory rate, heart rate, arterial blood gases (ABG) analysis and PaO2/FiO2 ratio. Other information collected at admission included impairment of consciousness (Kelly–Matthay scale) and severity of illness (SAPS II score) (Simplified Acute Physiology Score II). The ventilator settings, the changes in ABG after 2 h of NIV (17–19), the delay failure of NIV (20), the duration of NIV and the length of hospital stay were documented. The outcomes (intubation and other complications) at discharge or death were recorded. Inspiratory positive airway pressure (IPAP)
and expiratory positive airway pressure were set according to patient’s tolerance and to maintain a tidal volume of 6–8 ml/kg. All blood gas parameters were measured after two and 24 h of NIV and at discharge. A 6-month follow-up (including 180 day mortality) was done. Twenty-one patients were unavailable for follow-up visits. These were contacted by telephone to collect information relating to the variables documented during the inpatient period. Information was available for all patients. A comparison of outcomes in the groups was done. The flowchart of the study population is shown in Figure 1. The primary outcomes were the intubation and mortality rates. The secondary outcomes were: changes in ABG analysis 2 and 24 h after starting NIV, NIV duration and length of hospital stay. We also recorded 6 month mortality rate both by checking the hospital records or by telephone interview.
Statistical analysis Categorical variables were expressed as percentages and continuous variables as the mean SD. Categorical variables were compared between the two groups (patients aged 75 years or more and patients younger than 75) using the v2 and continuous variable using the regression analysis. Univariate logistic regression analysis and stepwise logistic regression were performed to determinate factors predicting failure of ventilation (hospital mortality). A p ≤ 0.05 was considered statistically significant. Statistical tests were performed using statistics software R-Project version 2.13.2.
Results During the 2-year study period, 207 patients who required non-invasive ventilatory support were admitted to our Respiratory Monitoring Units. One hundred and twenty-one of them (58.45%) were 75 or more years old and 86 (41.55%) younger.
Characteristics of very old patients In the 75 or older group, there were 62 males and 59 females. In this group, we collected data on daily activities, the presence of caregivers, residence in nursing home facilities, dyspnoea scale (MMRC) and use of drugs. The mean Barthel Index was 86.25 7.9 (moderate dependence). Among the 121 patients, 109 had a previous diagnosis of COPD with a mean FEV1% 37.8 10.7; 79 of them (65.2%) belonged to group D of GOLD combined COPD assessment (16) and had two or more COPD exacerbations annually. Fifteen of 121 (12.3%) had been admitted to the ICU previously. Sixty-two patients ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
NIV in elderly people
207 paents admied to Respiratory Monitoring Care Units for ARF due to COPD or AHF or both during the study period
121
paents
86 paents aged < 75 years
aged ≥ 75 years
NIV
NIV
ETI 13 paents 24 paents died (12 do not intubate order)
97 paents survived (80.16%)
ETI 10 paents; 9 paents died
76 paents survived (88.37%)
ETI = endotracheal intubaon Figure 1 The flowchart of the study population
(51.2%) used home oxygen therapy and fourteen (11.5%) used home mechanical ventilation. In the other patients, the diagnosis of COPD was confirmed by spirometry performed during the hospital stay. Twenty-two (18.2%) of the 121 very old patients had Charlson Comorbidity Index < 5. The most frequent comorbidities were chronic heart failure (31%), diabetes mellitus (22%), cerebrovascular diseases (18%), peripheral vascular diseases (16%), renal failure (14%) and haematological diseases (6%).
performed during hospitalisation. Sixty-nine (80.2%) of the 86 patients under 75 years of age had a Charlson Comorbidity Index < 5. The most frequent comorbidities were chronic heart failure (16%), diabetes mellitus (14%), cerebrovascular diseases (11%), peripheral vascular diseases (10%), renal failure (6%) and connective tissue disorders (5%). The baseline characteristics of older and younger patients are summarised in Table 1.
Young and old survival and functional status Characteristics of younger patients In the under 75 year group, 44 were males and 42 females. In the younger group, the same physiopathological and demographical data was collected. The mean Barthel Index was 92.3 4.1 (mild dependence). Of the 86 patients, 80 had a previous diagnosis of COPD with a mean FEV1% 36.9 11.5 and of these (67.4%) belonged to group D GOLD combined COPD assessment (16) and had 2 or more COPD exacerbations per year. Seven (9.3%) had previously ICU admission. Twenty-nine (33.7%) utilised home oxygen therapy and eight (9.3%) used home mechanical ventilation. In the remaining patients, the diagnosis of COPD was confirmed by spirometry ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
The median follow-up was 184 (range 179–194) days after hospital discharge. The number of patients who died in the very old group was 24/121 (19.8%). Of the very old patients, 13 of 121 (10.74%) were intubated. Twelve died. Twelve other patients were not intubated because of a ‘do not intubate’ advance directive. Of the 97 very old patients discharged alive from hospital, 39 (40.2%) were discharged at home and 58 (59.7%) into a nursing home. Forty-seven of the 97 hospital survivors died after hospital discharge within 6 month (range 27–181 days with median time of 128).The overall 3 month and 6 month mortalities were 29.4% and 38.8%, respectively.
1525
1526
NIV in elderly people
Table 1 Baseline demographic, social, functional and treatment patient characteristics according to age
Older patients
Primary caregivers Inpatient nursing home Any type of assistance Dyspnoea (MMRC) III-IV grade Drugs per day Comorbidities (Charlson index) Barthel Index score FEV1% COPD assessment D Number of exacerbation (≥ 2 per year) Previous ICU admission Home oxygen therapy Home mechanical ventilation
Younger patients
No. patients
Percentage of total%
No. patients
Percentage of total%
p-value
66/121 21/121 34/121 89/121 4.10 1.8 5.88 2.8 86.25 7.9 37.8 10.7 79/121
54.5 17.4 28.1 73.5
40.6 8.1 51.1 89.5
65.2
35/86 7/86 44/86 77/86 2.9 1.6 4.32 1.7 92.3 4.1 36.9 11.5 58/86
67.4
0.03 0.02 0.003 0.03 0.04 0.01 0.03 0.07 0.18
79/121 15/121 62/121 14/121
65.2 12.3 51.2 11.5
58/86 7/86 29/86 8/86
67.4 9.3 33.7 9.3
0.11 0.07 0.04 0.06
H-mortality
180-day_mortality
45.0% 38.8%
40.0% 35.0% 30.0% 25.0% 18.6%
20.0% Figure 2 The percentage of NIV failure in the two groups
15.0%
Among the younger patients, 10 of 86 (11.6%) were intubated. Nine died; one survived (7.69%). Sixteen of the 77 hospital survivors died within 6 months after hospital discharge (range 32–179 days with median time of 124 days). The overall 3 month and 6 month mortality were 14.8% and 18.6%, respectively. The percentage of NIV failure in the two groups is reported in Figure 2. The mortality rate (hospital mortality and 180 day mortality) is summarised in Figure 3.
10.0%
Comparison of hospital outcome between old and younger patients The very old patients had significantly longer NIV duration 69.0 27.0 h vs. 57.0 27.0 (p < 0.03) and hospital stay 11.6 3.8 vs. 8.4 1.4 days (p < 0.02). At admission, very old patients did not have a significantly lower level of consciousness (Kelly scale), paO2,paCO2 and PaO2/FiO2 ratio. The older group had a significantly lower pH (p < 0.001) and greater severity of illness as determined by SAPS II (p < 0.001). Hospital mortality was similar in the two groups, as was intubation rate (NIV failure).
19.8%
10.5%
5.0% 0.0% Under 75 Over 75 Figure 3 Mortality rate (hospital and 180 day mortality) in
the two groups
Survival after 180 days after discharge was significantly higher in the younger group (p < 0.01). Intubation and mortality were strictly related to the presence of comorbidities and the severity of the respiratory picture at admission (p < 0.001). Table 2 demonstrates comorbidities and physiological parameters measured during hospital stay. The NIV complications were very low (8%).The most common complications were skin lesions (from erythema to skin ulceration), eye irritation, claustrophobia and gastric distension. Serious complications were not observed. An internal protocol of skin lesion care and NIV complication prevention was strictly followed. In case of intolerance of a type of mask during NIV, it was promptly changed to another type (e.g. total face mask or oro-nasal mask). ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
NIV in elderly people
Table 2 Comorbidities and physiological parameters measured during hospital stay according to age
Group under 75 Mean + SD
Age 70 3 Male/Female 44/42 Comorbidities 69 (Charlson Comorbidity Index CCI
The use of non-invasive ventilation in very old patients with hypercapnic acute respiratory failure because of COPD exacerbation A. Nicolini,1 M. Santo,2 L. Ferrera,3 M. Ferrari-Bravo,4 C. Barlascini,5 A. Perazzo1
1
SUMMARY
What’s known
Aims: We prospectively enrolled 207 patients (121 were 75 or older and 86 younger than 75) who were admitted to three Respiratory Monitoring Units. The primary outcomes were intubation and mortality rates; the secondary outcomes were changes in arterial blood gases analysis, non-invasive ventilation (NIV) duration and length of hospital stay. Results: Hospital mortality was similar in the two groups, as were intubation rates. The proportion who died in the very old patient group was 19.8% (24/121) vs. 10.4% (9/86) in the younger group. Intubation rate was 10.7% (13/121) in the very old patient group and 11.6% (10/86) in the younger group. The presence of comorbidities, the severity of illness (SAPS II), the level of consciousness, NIV failure (intubation), absolute value of pH prior to NIV, as well as the changes in pH and paCO2 and PaO2/FiO2 after 2 h of NIV, were the variables associated with higher mortality. Very old patients had significantly higher NIV duration than younger patients (69.0 47.0 vs. 57.0 27.0 h) (p ≤ 0.03) and hospital stays (11.6 3.8 vs. 8.4 1.4) (p ≤ 0.02). Conclusions: The use of NIV in very old patients was effective in many cases. Endotracheal intubation after NIV failure was not efficacious in either group.
Introduction The use of non-invasive ventilation (NIV) as alternative to endotracheal intubation in patients with chronic obstructive pulmonary disease (COPD) exacerbations or acute cardiogenic pulmonary oedema is supported by randomised controlled trials and metaanalyses (1). More recently, NIV has been employed successfully in settings outside the intensive care unit (ICU), including hospital wards (1,2). In these settings, NIV showed a success rate for COPD patients similar to that reported in ICU (1,3,4). The choice of NIV aims to avoid complications, particularly in patients with complicated pathological situations (5). Patients aged 75 years or older, also referred to as ‘very old patients,’ are potentially ‘good candidates’ for less invasive management (6). The proportion of elderly persons among hospitalised patients, including ICU admissions, is rapidly increasing in developed countries. Very old patients represent 10–15% of ICU admissions (6–8). The management of critical respiratory illness in old patients is of increasing importance.
Non-invasive ventilation (NIV) is an alternative choice to intubation in hypercapnic respiratory failure due to exacerbations of COPD. It avoids complications associated with intubation. Very old patients are good candidates for NIV.
What’s new This prospective study has demonstrated that NIV in very old patients achieved a high rate of success. Endotracheal intubation after NIV failure is of questionable value in very old COPD patients. NIV success is inversely related to the number and the severity of co-morbidities, level of consciousness and directly related to an early resolution of the respiratory distress.
Furthermore, NIV is utilised for the respiratory support of patients with a ‘do not intubate’ advance directive (9–12). Certainly, ‘do not intubate’ cannot be considered as an indication for NIV; however, the use of NIV as a palliative measure is increasing (6). The results of short-term use are promising, but specific data for long-term outcomes and mortality are limited (6,13–15). These considerations prompted us to study the value of NIV in very old patients (aged ≥ 75 years) presenting with acute respiratory failure (ARF). We collected data in the hospital and 6 months after discharge. A younger group (< 75 years of age) was used for comparison.
Respiratory Medicine Unit, ASL4 Chiavarese, Sestri Levante, Italy 2 Emergency Medicine Department, ASL4 Chiavarese, Lavagna, Italy 3 Respiratory Diseases Department, Villa Scassi Hospital, Genova, Italy 4 Hygiene and Public Health Department, ASL4 Chiavarese, Chiavari, Italy 5 Forensic Medicine, ASL4 Chiavarese, Sestri Levante, Italy Correspondence to: Nicolini Antonello, Respiratory Medicine Unit, Via Terzi 43 – General Hospital, 16039 Sestri Levante, Italy Tel.: + 00390185329145 Fax: + 00390185329935 Email: antonello.nicolini@ fastwebnet.it
Disclosure None.
Methods The study was conducted in the Respiratory Monitoring Unit (RMU) and Emergency Medicine Unit of ASL4 Chiavarese, Italy and the RMU of the Respiratory Diseases Department of Villa Scassi Hospital, ASL3 Genovese, Italy, from May 2011 to June 2013.
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529. doi: 10.1111/ijcp.12484
1523
1524
NIV in elderly people
The study was approved by the Ethics Committee of ASL4 Chiavarese (no. 502 – 11/9/2010). These three units each have four non-invasive monitored beds and admit patients with severe respiratory failure who need NIV. We prospectively enrolled 207 patients with exacerbations of COPD, severe hypercapnic respiratory failure (ARF) and moderate to severe acidosis. One hundred and twenty-one of the 207 patients were 75 or older; 86 were younger than 75 years. All the patients were treated with non-invasive ventilation. One hundred and ninety-five patients gave written informed consent. In the other 12 patients who were unable to give written informed consent, the closest family member gave written informed consent. Diagnosis of COPD was based on the presence of airflow obstruction observed in previous pulmonary function tests. The severity of disease was based upon GOLD criteria (16). The RMUs had ventilators specifically designed for NIV (Philips Respironics Vision BiPap, Philips Respironics V60) and invasive ventilation (Philips Respironics V 200 Esprit, Versamed Ivent 201). Thus, it was possible to switch from NIV to invasive ventilation at any moment. Criteria for starting NIV as well as criteria of exclusion were previously described (17–19). Furthermore, patients enrolled for NIV did not present any criteria for emergency intubation (e.g. respiratory pauses, agitation requiring sedation, hemodynamic instability, systolic blood pressure < 90 mmHg or heart rate less than 40 beats/min). Pressure support ventilation was the preferred mode delivered by oro-nasal (full-face) or total face mask. We collected demographical and daily life data (age, sex, dyspnoea according to the classification of the MMRC, number of drugs taken per day, daily life limitations measure according to the Barthel test) and previous medical history (number of comorbidities, Charlson index, oxygen therapy, home nocturnal or diurnal ventilation or both, previous admissions to RMU or Intensive Care, and lung function. Moreover, the reasons for admission to our units and the severity of the respiratory signs were recorded. These included respiratory rate, heart rate, arterial blood gases (ABG) analysis and PaO2/FiO2 ratio. Other information collected at admission included impairment of consciousness (Kelly–Matthay scale) and severity of illness (SAPS II score) (Simplified Acute Physiology Score II). The ventilator settings, the changes in ABG after 2 h of NIV (17–19), the delay failure of NIV (20), the duration of NIV and the length of hospital stay were documented. The outcomes (intubation and other complications) at discharge or death were recorded. Inspiratory positive airway pressure (IPAP)
and expiratory positive airway pressure were set according to patient’s tolerance and to maintain a tidal volume of 6–8 ml/kg. All blood gas parameters were measured after two and 24 h of NIV and at discharge. A 6-month follow-up (including 180 day mortality) was done. Twenty-one patients were unavailable for follow-up visits. These were contacted by telephone to collect information relating to the variables documented during the inpatient period. Information was available for all patients. A comparison of outcomes in the groups was done. The flowchart of the study population is shown in Figure 1. The primary outcomes were the intubation and mortality rates. The secondary outcomes were: changes in ABG analysis 2 and 24 h after starting NIV, NIV duration and length of hospital stay. We also recorded 6 month mortality rate both by checking the hospital records or by telephone interview.
Statistical analysis Categorical variables were expressed as percentages and continuous variables as the mean SD. Categorical variables were compared between the two groups (patients aged 75 years or more and patients younger than 75) using the v2 and continuous variable using the regression analysis. Univariate logistic regression analysis and stepwise logistic regression were performed to determinate factors predicting failure of ventilation (hospital mortality). A p ≤ 0.05 was considered statistically significant. Statistical tests were performed using statistics software R-Project version 2.13.2.
Results During the 2-year study period, 207 patients who required non-invasive ventilatory support were admitted to our Respiratory Monitoring Units. One hundred and twenty-one of them (58.45%) were 75 or more years old and 86 (41.55%) younger.
Characteristics of very old patients In the 75 or older group, there were 62 males and 59 females. In this group, we collected data on daily activities, the presence of caregivers, residence in nursing home facilities, dyspnoea scale (MMRC) and use of drugs. The mean Barthel Index was 86.25 7.9 (moderate dependence). Among the 121 patients, 109 had a previous diagnosis of COPD with a mean FEV1% 37.8 10.7; 79 of them (65.2%) belonged to group D of GOLD combined COPD assessment (16) and had two or more COPD exacerbations annually. Fifteen of 121 (12.3%) had been admitted to the ICU previously. Sixty-two patients ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
NIV in elderly people
207 paents admied to Respiratory Monitoring Care Units for ARF due to COPD or AHF or both during the study period
121
paents
86 paents aged < 75 years
aged ≥ 75 years
NIV
NIV
ETI 13 paents 24 paents died (12 do not intubate order)
97 paents survived (80.16%)
ETI 10 paents; 9 paents died
76 paents survived (88.37%)
ETI = endotracheal intubaon Figure 1 The flowchart of the study population
(51.2%) used home oxygen therapy and fourteen (11.5%) used home mechanical ventilation. In the other patients, the diagnosis of COPD was confirmed by spirometry performed during the hospital stay. Twenty-two (18.2%) of the 121 very old patients had Charlson Comorbidity Index < 5. The most frequent comorbidities were chronic heart failure (31%), diabetes mellitus (22%), cerebrovascular diseases (18%), peripheral vascular diseases (16%), renal failure (14%) and haematological diseases (6%).
performed during hospitalisation. Sixty-nine (80.2%) of the 86 patients under 75 years of age had a Charlson Comorbidity Index < 5. The most frequent comorbidities were chronic heart failure (16%), diabetes mellitus (14%), cerebrovascular diseases (11%), peripheral vascular diseases (10%), renal failure (6%) and connective tissue disorders (5%). The baseline characteristics of older and younger patients are summarised in Table 1.
Young and old survival and functional status Characteristics of younger patients In the under 75 year group, 44 were males and 42 females. In the younger group, the same physiopathological and demographical data was collected. The mean Barthel Index was 92.3 4.1 (mild dependence). Of the 86 patients, 80 had a previous diagnosis of COPD with a mean FEV1% 36.9 11.5 and of these (67.4%) belonged to group D GOLD combined COPD assessment (16) and had 2 or more COPD exacerbations per year. Seven (9.3%) had previously ICU admission. Twenty-nine (33.7%) utilised home oxygen therapy and eight (9.3%) used home mechanical ventilation. In the remaining patients, the diagnosis of COPD was confirmed by spirometry ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
The median follow-up was 184 (range 179–194) days after hospital discharge. The number of patients who died in the very old group was 24/121 (19.8%). Of the very old patients, 13 of 121 (10.74%) were intubated. Twelve died. Twelve other patients were not intubated because of a ‘do not intubate’ advance directive. Of the 97 very old patients discharged alive from hospital, 39 (40.2%) were discharged at home and 58 (59.7%) into a nursing home. Forty-seven of the 97 hospital survivors died after hospital discharge within 6 month (range 27–181 days with median time of 128).The overall 3 month and 6 month mortalities were 29.4% and 38.8%, respectively.
1525
1526
NIV in elderly people
Table 1 Baseline demographic, social, functional and treatment patient characteristics according to age
Older patients
Primary caregivers Inpatient nursing home Any type of assistance Dyspnoea (MMRC) III-IV grade Drugs per day Comorbidities (Charlson index) Barthel Index score FEV1% COPD assessment D Number of exacerbation (≥ 2 per year) Previous ICU admission Home oxygen therapy Home mechanical ventilation
Younger patients
No. patients
Percentage of total%
No. patients
Percentage of total%
p-value
66/121 21/121 34/121 89/121 4.10 1.8 5.88 2.8 86.25 7.9 37.8 10.7 79/121
54.5 17.4 28.1 73.5
40.6 8.1 51.1 89.5
65.2
35/86 7/86 44/86 77/86 2.9 1.6 4.32 1.7 92.3 4.1 36.9 11.5 58/86
67.4
0.03 0.02 0.003 0.03 0.04 0.01 0.03 0.07 0.18
79/121 15/121 62/121 14/121
65.2 12.3 51.2 11.5
58/86 7/86 29/86 8/86
67.4 9.3 33.7 9.3
0.11 0.07 0.04 0.06
H-mortality
180-day_mortality
45.0% 38.8%
40.0% 35.0% 30.0% 25.0% 18.6%
20.0% Figure 2 The percentage of NIV failure in the two groups
15.0%
Among the younger patients, 10 of 86 (11.6%) were intubated. Nine died; one survived (7.69%). Sixteen of the 77 hospital survivors died within 6 months after hospital discharge (range 32–179 days with median time of 124 days). The overall 3 month and 6 month mortality were 14.8% and 18.6%, respectively. The percentage of NIV failure in the two groups is reported in Figure 2. The mortality rate (hospital mortality and 180 day mortality) is summarised in Figure 3.
10.0%
Comparison of hospital outcome between old and younger patients The very old patients had significantly longer NIV duration 69.0 27.0 h vs. 57.0 27.0 (p < 0.03) and hospital stay 11.6 3.8 vs. 8.4 1.4 days (p < 0.02). At admission, very old patients did not have a significantly lower level of consciousness (Kelly scale), paO2,paCO2 and PaO2/FiO2 ratio. The older group had a significantly lower pH (p < 0.001) and greater severity of illness as determined by SAPS II (p < 0.001). Hospital mortality was similar in the two groups, as was intubation rate (NIV failure).
19.8%
10.5%
5.0% 0.0% Under 75 Over 75 Figure 3 Mortality rate (hospital and 180 day mortality) in
the two groups
Survival after 180 days after discharge was significantly higher in the younger group (p < 0.01). Intubation and mortality were strictly related to the presence of comorbidities and the severity of the respiratory picture at admission (p < 0.001). Table 2 demonstrates comorbidities and physiological parameters measured during hospital stay. The NIV complications were very low (8%).The most common complications were skin lesions (from erythema to skin ulceration), eye irritation, claustrophobia and gastric distension. Serious complications were not observed. An internal protocol of skin lesion care and NIV complication prevention was strictly followed. In case of intolerance of a type of mask during NIV, it was promptly changed to another type (e.g. total face mask or oro-nasal mask). ª 2014 John Wiley & Sons Ltd Int J Clin Pract, December 2014, 68, 12, 1523–1529
NIV in elderly people
Table 2 Comorbidities and physiological parameters measured during hospital stay according to age
Group under 75 Mean + SD
Age 70 3 Male/Female 44/42 Comorbidities 69 (Charlson Comorbidity Index CCI
