Outcomes of long-term ventilator patients: a descriptive study.

BACKGROUND: Long-term ICU patients who require prolonged mechanical ventilation are a growing segment of the in-hospital population. Despite recognition that this population is costly to care for no systematic research has been done on the characteristics, outcomes, and disposition of these patients after they leave the hospital. OBJECTIVE: To describe clinical and sociodemographic characteristics and outcomes of ICU patients who require long-term (5 days or more) mechanical ventilation while in the hospital. METHODS: A prospective, longitudinal descriptive design was used to study 57 ICU patients who required 5 days or more of continuous mechanical ventilation while in the hospital. Clinical and sociodemographic data were collected at the time of enrollment. Patients were followed up for up to 6 months after discharge from the hospital to ascertain disposition and morality. RESULTS: On average, patients had a hospital stay of almost 6 weeks and required mechanical ventilation for approximately 4 weeks; 43.9% of the patients died in the hospital. None of the patients discharged from the hospital were able to return home initially without assistance. By 6 months after discharge, more than 50% of the original sample and died, 9% resided in an institution, and 33% were living at home. CONCLUSIONS: A large percentage of ICU patients who require 5 days or more of mechanical ventilation die in the hospital, and many of those who live spend considerable time in an extended-care facility before they are discharged to their homes. These likely outcomes of patients who require long-term ventilation should be discussed with patients and their families to assist them in making informed decisions.

Long-term intramuscular electrical activation of the phrenic nerve: safety and reliability.

The safety and reliability of a system for long-term intramuscular electrical activation of the phrenic nerve was evaluated in seven dogs. In this system, electrodes are implanted bilaterally into the diaphragm without directly contacting the phrenic nerve using a laparoscope to direct placement. Five dogs underwent chronic bilateral intramuscular diaphragm stimulation (IDS) for 61 to 183 days at stimulus parameters selected to evoke at least 120% of the animal's basal ventilation. Two dogs maintained as controls did not undergo chronic stimulation. The safety and reliability of the system was evaluated in terms of tissue responses to the electrode, alterations in diaphragm muscle, pulmonary function, electrode reliability, and cardiac activation. (The efficacy of long-term intramuscular activation of the phrenic nerve is addressed in a companion paper.) No adverse responses to the electrode or stimulation were found. The histochemistry of chronically stimulated diaphragm suggested transformation towards type I (oxidative metabolism) muscle fibers. Two IDS electrodes dislodged out of a total of 32 IDS electrodes implanted. Both electrodes dislodged within seven days of implant. All IDS electrodes had stable and repeatable recruitment properties. No IDS electrode mechanical failures were found and no electrode corrosion was observed. We conclude from these experiments that intramuscular activation of the phrenic nerve will present a minimal risk to human patients who are good candidates for clinical studies using this technique.

Long-term intramuscular electrical activation of the phrenic nerve: efficacy as a ventilatory prosthesis.

The efficacy of a system for long-term intramuscular activation of the phrenic nerve as a ventilatory prosthesis was evaluated in seven dogs. (The safety and reliability of this system is addressed in a companion paper.) Five dogs underwent chronic bilateral intramuscular diaphragm stimulation (IDS) for 61 to 183 days at stimulus parameters selected to evoke at least 120% of the animal's basal ventilation. Two dogs maintained as controls did not undergo chronic stimulation. The ability of IDS to provide long-term ventilation without diaphragm fatigue was evaluated in terms of the ventilatory capacity of IDS, the effects of chronic IDS on diaphragm contractile properties, and the phrenic nerve recruitment properties of chronic IDS electrodes. Hemidiaphragms with electrodes placed within 2 cm of the phrenic nerve trunk could be completely activated by 25 mA pulses having a 100 microsecond pulse width. The tidal volume evoked by IDS in this study was 167% (+/- 48 s.d.) of that required for full-time basal ventilation without diaphragm fatigue. Evoked tidal volume increased after 8 to 9 weeks of chronic IDS for stimulus pulse intervals longer than 50 mS. Electrode recruitment properties were stable for both active and passive implanted electrodes. We conclude from these studies that with properly placed electrodes IDS is capable of providing reliable full-time ventilatory support without fatiguing the diaphragm.

Respiratory function, cognitions, and panic in chronic obstructive pulmonary patients.

The present study investigated the relationship between respiratory function, catastrophic thoughts about anxiety, and panic in 48 Chronic Obstructive Pulmonary Disease (COPD) patients. During a routine office visit which included respiratory function tests (Forced Vital Capacity, FVC; Forced Expiratory Volume--first second, FEV1) patients completed a battery of questionnaires which assessed history of panic, days with shortness of breath, general activity level, agoraphobic cognitions, perception of bodily sensations, anxiety and depression. Thirty-seven percent of the sample reported experiencing a panic attack. Subjects showed a significant impairment in respiratory functioning. Patients with a history of panic did not differ from those who had not experienced panic on demographic, physiologic, or activity variables. Patients who experienced panic reported significantly more agoraphobic cognitions and greater concern with bodily sensations than did patients who did not experience panic.

Taking the laparoscope to the laboratory for ventilatory research.

Chronic ventilatory support by phrenic nerve stimulation has been accomplished, but potential permanent phrenic nerve injury has limited its therapeutic use. Prior work from the authors' laboratory has demonstrated that direct muscle stimulation can be safely performed for prolonged periods. Thus, in 1983, a laparoscopic technique was devised for the direct implantation of diaphragm pacing electrodes in the canine. During the subsequent four years, 16 animals (10 acute and 6 chronic) have undergone diaphragm pacing. Under general anesthesia, diaphragm pacing electrodes loaded into 19-gauge needles were implanted under direct laparoscopic visualization in both hemidiaphragms. In each animal, tidal volume, transdiaphragmatic pressure, arterial blood gas, impedence, cardiac activation, and histology of electrode placement site have been evaluated. Chronically paced animals were subjected to daily 8- to 24-hr per day pacing. The longest paced animal has been 179 days with continuous 24-hour per day pacing. In both acute and chronic experiments, maximal evoked tidal volume has approached 400 per cent of resting tidal volume. Cardiac ectopy has not occurred. This work has demonstrated the safety and feasibility of direct muscular diaphragm activation using the laparoscope. It has provided the basis for embarking upon a clinical study of this modality as a method of chronic ventilatory support. The laparoscopic technique of electrode placement may have a variety of other potential applications.

The treatment center: a model for competitive clinical practice.

The treatment center model described in this article was developed with the intent of providing a high technology practice environment that would attract patients to an academic center for their medical care. It was also intended to create an environment that would result in the retention of the best clinicians within the academic framework. Appropriate goals and incentives had to be determined by both the hospital and the professional groups involved. Many preconceived ideas regarding traditional academic structures needed to be revised to implement this program.

Electrical activation of the diaphragm.

The physiology of electrical activation of the diaphragm is reviewed, with emphasis upon the respiratory mechanics of the diaphragm and its action upon other respiratory structures. The application of diaphragm pacing is discussed in terms of candidate selection, evaluation, and its future as a treatment modality for respiratory insufficiency.

The impact of age on utilization of intensive care resources.

The impact of age on admission practices and pattern of care were examined in 599 admissions to a medical intensive care unit (MICU) and 290 patients on the conventional medical care divisions of the same hospital. Four age groups were compared: under 55, 55 to 64, 65 to 74, and 75 years of age and over. Severity of illness and prior health were assessed using the Acute Physiology Score (APS) and the Chronic Health Evaluation (CHE) instruments. Resource utilization was assessed using the Therapeutic Intervention Scoring System (TISS) and hospital charges. Patients 65 years of age and over comprised 48% of the MICU sample. The distribution of CHE was different among the four groups. Twenty-one percent of patients under 55 years of age had no prior chronic illness, as compared to less than 8% of older patients. The APS at admission was similar for all age groups, as was admission, daily, and total TISS. Hospital survival declined with age from 85% to 70%, while the likelihood of being designated do not resuscitate (DNR) increased from 10% to 24%. Differences in hospital survival disappeared when controlling for severity of illness and prior health, but differences in DNR status did not. Still, elderly DNR patients received as much resources as younger DNR patients and this was more than non-DNR patients. The sample of patients treated on conventional medical divisions had age distribution similar to the MICU sample. There was some evidence that admission APS (median, 5, 5, 6, 6, respectively, P = .055) and maximum APS (median, 5, 5, 7, 8, respectively, P = .023) differed slightly across age groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparative effects of aminophylline on diaphragm and cardiac contractility.

The mechanisms by which aminophylline increases inspiratory muscle contractility are unclear. The present study compared the effects of aminophylline on cardiac as well as on diaphragm contractility and examined the interaction of aminophylline with verapamil, a calcium channel-blocking agent, on both types of muscle. Experiments were performed in mongrel dogs anesthetized with pentobarbitone. Diaphragm contractility was assessed from transdiaphragmatic pressure (Pdi) developed during supramaximal electrical stimulation of the cervical phrenic rootlets, and cardiac contractility was assessed from peak left ventricular (LV) pressure and its rate of rise (dPv/dt). Measurements were made before and after each of 3 sequential IV infusions of aminophylline (6 mg/kg) and subsequent IV infusions of verapamil (0.1 mg/kg bolus and 0.02 mg/kg/min for 5 to 7 min.). Transient decreases in Pdi were frequently observed immediately after aminophylline infusion in association with decreases in mean arterial blood pressure. With recovery of mean blood pressure, Pdi increased above baseline values. Aminophylline increased Pdi in a dose-dependent fashion over the entire frequency range studied (1 to 40 Hz). Aminophylline increased the rate of rise of Pdi (dPd/dt) without affecting the period over which pressure was developed or dissipated during single twitches. Aminophylline increased both peak LV pressure and dPv/dt. The magnitude of the cardiac response was greater than the diaphragmatic response. Subsequent verapamil infusion completely reversed the effects of aminophylline on LV contractility but had only a small effect on diaphragm contractility.(ABSTRACT TRUNCATED AT 250 WORDS)

Conditioning of the diaphragm with phrenic nerve stimulation after prolonged disuse.

We studied the effect of electrical stimulation of the phrenic nerve on the force frequency relationship of the disused diaphragm. A high quadriplegic who had been totally ventilator dependent for 6 months following a C2 fracture received bilateral phrenic nerve stimulators. During a 6-wk period of conditioning by electrical stimulation, the force of diaphragm contraction was assessed by measurement of transdiaphragmatic pressures during stimulation of each nerve over a range of frequencies. Tidal volume as well as rib cage and abdominal motion were studied. There was an upward shift of the force frequency relationship of the diaphragm over the 4-month period of phrenic nerve pacing using repetitive stimulus trains of 14 to 28 Hz. This improvement appeared to plateau at about 11 wk. The increase in contractility was accompanied by a progressive diminution in the stimulus frequency at which fusion of the contraction occurred. The disused diaphragm, like other skeletal muscle, may be conditioned with electrical stimulation.

Diaphragm activation with intramuscular stimulation in dogs.

We studied in 10 supine anesthetized dogs diaphragm contraction produced by electrical activation with intramuscular electrodes surgically implanted in the ventral surface of the diaphragm and compared this with activation of the ipsilateral phrenic nerve (C5, 6, and 7) before it entered the thorax. Repetitive 40-Hz pulse trains with supramaximal current stimulus were used after hyperventilation of the animals to apnea. A single intramuscular electrode within 1 to 2 cm of the site of phrenic nerve entry into the diaphragm produced a mean transdiaphragmatic pressure of 12.0 cm H2O +/- 0.97 SE and mean tidal volume of 0.27 L +/- 0.04 SE. Mean values observed with phrenic nerve stimulation were not statistically different, and both electrode systems produced equivalent outward abdominal motion and upper rib cage paradox, as monitored by inductive plethysmography. There was no difference in gas exchange during stimulation with a single hemidiaphragm electrode and mechanical ventilation compared at the same tidal volume and respiratory rate. Blockade of neuromuscular transmission with curare eliminated intramuscular and phrenic nerve stimulation proportionately, suggesting that activation of the diaphragm is dependent in both cases on the phrenic nerve. This technique does not entail manipulation of the phrenic nerve and may have clinical application as an alternative technique for diaphragm pacing.

Fatigue of the mammalian diaphragm in vitro.

Diaphragm fatigue was studied in innervated diaphragm strips from 63 Sprague-Dawley rats. The experiments examined 1) the effect on the rate of diaphragmatic fatigue of increases in the diaphragm's duty cycle, i.e., the ratio of the period of diaphragmatic contraction (Ti) to the duration of a cycle of contraction and rest (Ttot) and 2) the possibility that impaired neural transmission contributed to the fatigue process. Alterations in the duty cycle of the diaphragm were simulated by varying the pattern of electrical stimuli applied cyclically to the phrenic nerve. Fatigue was assessed from the rate of fall of isometric tension when the muscle was made to contract 90 times/min. The contribution of neural element fatigue was assessed by comparing the tension during phrenic nerve stimulation to the tension developed when the muscle was stimulated directly. Increasing the duty cycle (Ti/Ttot) from 25 to 50 to 75% increased the rate of diaphragmatic fatigue progressively. Holding Ti/Ttot constant at 75%, while varying Ti and Ttot, did not affect the rate of fatigue. Increases in duty cycle appear to increase the rate of fatigue by increasing the number of times the contractile process was activated. In fatigued muscle strips diaphragmatic tension was greater in directly stimulated muscle than in muscle strips activated via the phrenic nerve. The results indicate that 1) when the breathing action of the diaphragm is simulated in vitro, increases in duty cycle accelerate the fatigue process and 2) failure of transmission of phrenic impulses to diaphragmatic muscle cells contributes to the fall in tension during fatigue.

Respiratory responses in reversible diaphragm paralysis.

The effects of diaphragm paralysis on respiratory activity were assessed in 13 anesthetized, spontaneously breathing dogs studied in the supine position. Transient diaphragmatic paralysis was induced by bilateral phrenic nerve cooling. Respiratory activity was assessed from measurements of ventilation and from the moving time averages of electrical activity recorded from the intercostal muscles and the central end of the fifth cervical root of the phrenic nerve. The degree of diaphragm paralysis was evaluated from changes in transdiaphragmatic pressure and reflected in rib cage and abdominal displacements. Animals were studied both before and after vagotomy breathing O2, 3.5% CO2 in O2, or 7% CO2 in O2. In dogs with intact vagi, both peak and rate of rise of phrenic and inspiratory intercostal electrical activity increased progressively as transdiaphragmatic pressure fell. Tidal volume decreased and breathing frequency increased as a result of a shortening in expiratory time. Inspiratory time and ventilation were unchanged by diaphragm paralysis. These findings were the same whether O2 or CO2 in O2 was breathed. After vagotomy, no significant change in phrenic or inspiratory intercostal activity occurred with diaphragm paralysis in spite of increased arterial CO2 partial pressure. Ventilation and tidal volume decreased significantly, and respiratory timing was unchanged. These results suggest that mechanisms mediated by the vagus nerves account for the compensatory increase in respiratory electrical activity during transient diaphragm paralysis. That inspiratory time is unchanged by diaphragm paralysis whereas the rate or rise of phrenic nerve activity increases suggest that reflexes other than the Hering-Breuer reflex contribute to the increased respiratory response.