The Authors: Ahmed BaHammam is Professor of Medicine, Director of the University Sleep Disorders Center, King Saud University and Consultant, Pulmonary, Sleep and Critical Care Medicine. Research interests include non-invasive ventilation, obesity hypoventilation, sleep in critically ill patients and patients with acute coronary syndrome, oxidative stress in sleep disordered breathing and the effects of fasting on sleep. Suhaila Al-Jawder is Assistant Professor in Arabian Gulf University and Consultant, Pulmonary and Sleep Medicine at Salmaniya Medical Complex, and University Sleep Disorders Center, King Saud University. Research interests include hypoventilation syndromes, non-invasive modes of ventilation and sleep-related breathing disorders in cardiovascular diseases.
SERIES EDITOR: AMANDA J PIPER
Ahmed S. BaHammam, University Sleep Disorders Center, College of Medicine, Department of Medicine, King Saud University, Box 225503, Riyadh 11324, Saudi Arabia. Email: firstname.lastname@example.org; email@example.com
Morbid obesity adversely affects respiratory physiology, leading to reduced lung volumes, decreased lung compliance, ventilation perfusion mismatch, sleep-disordered breathing and the impairment of ventilatory control, and neurohormonal and neuromodulators of breathing. Therefore, morbidly obese subjects are at increased risk of various pulmonary complications that can present either acutely or chronically. Respiratory failure is one of the most common pulmonary complications related to morbid obesity. Both acute hypoxaemic and hypercapnic respiratory failure are more common among obese patients. The management pathway of respiratory failure depends, to a large extent, on the underlying cause, primarily due to the diversity of the underlying triggering diseases, the pathophysiology and the prognosis associated with each disease. Morbidly obese patients with hypoventilation have an increased risk of acute hypercapnic respiratory failure. Early diagnosis of this disorder and the application of non-invasive ventilation in this group of patients have been shown to improve respiratory parameters, decrease the need for invasive mechanical ventilation and improve survival. Invasive ventilation remains the last life-saving procedure in patients with respiratory failure who do not respond to non-invasive measures. However, due to the abnormal respiratory physiology in obese patients, special precautions are required during intubation, mechanical ventilation and weaning.
Obesity is an international epidemic that is associated with multiple health problems and has been increasing worldwide over the last few decades.1 Recent data indicate that the prevalence of morbid obesity is increasing significantly.2 The prevalence of obesity among intensive care unit (ICU) patients ranges from 9% to 26%, and the prevalence of morbid obesity, which is a known risk factor for metabolic syndrome, diabetes mellitus and cardiovascular diseases,3 ranges from 1.4% to 7%.4 Obesity also contributes to several sleep-related breathing disorder (SBD), including obstructive sleep apnoea (OSA) and obesity hypoventilation syndrome (OHS).5 Furthermore, the respiratory system is adversely affected at different levels by excess weight. The observed compromise of lung mechanics manifests at rest, with exertion, during acute illnesses and postoperatively.6 Thus, morbid obesity puts patients at an increased risk of developing acute or chronic respiratory failure. In these patients, comorbid respiratory and cardiac conditions, such as exacerbation of bronchial asthma, chronic obstructive pulmonary disease (COPD) or heart failure, can easily progress into acute hypoxaemic or hypercapnic respiratory failure.7 The diagnosis and treatment of respiratory complications in obese subjects, especially when they are critically ill, represent a new challenge both for health-care providers and to the economics of health-care services.8,9 One of these challenges is the management of acute respiratory failure (ARF) in obesity. In general, the data addressing the management of ARF in obese patients are limited. Therefore, it is important to develop a systematic approach for the assessment and management of obese patients with ARF utilizing respiratory-assisting devices guided by the available data. Creating a protocol for the proper utilization and application of non-invasive ventilation (NIV) in the acute care setting will enhance the success of this therapeutic modality and spare patients the need for invasive mechanical ventilation. This review first briefly discusses the pulmonary physiological changes of obesity that contribute to respiratory failure, and then provides a more detailed discussion of the assessment and management of both hypoxaemic and hypercapnic (ventilatory) ARF in morbidly obese patients based on the available evidence.
PULMONARY PHYSIOLOGY IN MORBID OBESITY
Obesity affects the respiratory system and ventilatory function at different levels. The alterations in the respiratory system in obesity can lead to the development of acute and chronic respiratory failure, SBD, and postoperative pulmonary complications. The pathophysiological respiratory changes in obesity must be recognized to understand the therapeutic implications of these changes and to help avoid potential pulmonary complications. The respiratory system can be affected by obesity in the following ways:
Lung volumes and mechanics
Obese subjects tend to have a lower functional residual capacity (FRC), expiratory reserve volume and total lung capacity compared with normal weight subjects.10 Furthermore, forced vital capacity, forced expiratory volume in 1 s, maximum voluntary ventilation and forced mid-expiratory flow are also significantly reduced in extremely obese subjects.11 The above changes are more prominent when patients are in the supine position due to the impedance of diaphragm movement by the abdomen.7 Airway closure during expiration may create an intrinsic auto-positive and expiratory pressure (PEEP).12 Breathing against auto-PEEP increases work of breathing (WOB), particularly in the supine position.13
Morbidly obese subjects experience an increased WOB even during quiet breathing, with a direct relationship between the increased WOB and partial arterial carbon dioxide concentration (PaCO2) levels.14 The WOB is further increased in obese subjects by the increased upper airway resistance. The peri-pharyngeal fat distribution in obesity reduces the pharyngeal cross-sectional area and contributes to the complete or partial obstruction of the upper airway.15 This mechanical obstruction explains both why SBD is common in obese subjects and why weight loss results in an increased pharyngeal cross-sectional area.16 To reduce the WOB, obese subjects adopt a special breathing pattern with reduced tidal volumes (VT) and higher respiratory rates compared with normal weight controls, resulting in an increase in dead space.17,18 Moreover, the significant reduction in FRC in morbidly obese subjects, particularly those with a high waist-to-hip ratio (central obesity), causes airway closure within the range of tidal breathing, resulting in significant pulmonary ventilation perfusion (V/Q) defects, hypoxaemia and poor gas exchange.19
The earlier data suggest that obesity generally compromises lung volume and mechanics, thereby contributing to hypoxaemia. Moreover, the earlier changes imply a decreased ventilatory reserve in obese subjects, with a tendency for acute ventilatory failure when additional workloads are imposed by new acute insults, such as infection or lung congestion.18
Studies have shown that the ventilation and perfusion of the lungs is unevenly distributed in obese subjects, which is more obvious in individuals with reduced expiratory reserve volume.20 Similar to normal weight subjects, obese persons maintain good perfusion of the lower dependent lung zones; however, due to closure of the small airways, ventilation is preferentially distributed to the upper lung zones, which leads to V/Q mismatch and hypoxaemia.21 These changes are more obvious in the supine position and during sleep.22 V/Q mismatch and changes in lung volumes make hypoxaemia common in both eucapnic obesity and OHS. However, hypoxaemia is usually worse and more sustained in OHS patients and occurs during both wakefulness and sleep.23 Sustained hypoxaemia, particularly in obese patients with OHS, predisposes them to pulmonary hypertension and congestive heart failure.24
In eucapnic obese subjects, both the maximal inspiratory and expiratory pressures are normal.25 However, in obese subjects with OHS, respiratory muscle strength can be half the normal level.26 In addition, the increased WOB and oxygen consumption by respiratory muscles causes the respiratory muscles in obese subjects to become fatigued and to have decreased endurance.27
Central ventilatory drive
To meet the increased rates of oxygen consumption and carbon dioxide production, obese subjects need to increase their minute ventilation to retain good oxygenation and adequate alveolar ventilation.13 In general, the ventilatory drive in eucapnic morbidly obese subjects is higher than that for normal weight subjects.13,28 Nevertheless, a subgroup of morbidly obese subjects has an impaired ventilatory response to hypoxaemia and hypercapnia.13 This subgroup fails to meet the increased ventilatory demands associated with their obesity, which increases their risk of developing awake hypoventilation and hypercapnia, and progressing into hypercapnic (ventilatory) ARF.5 An earlier study of OHS patients demonstrated that minute ventilation decreased by 21% during non-rapid eye movement sleep and 39% during rapid eye movement sleep compared with values recorded while awake.29 A genetic predisposition to hypoventilation was initially proposed; however, later studies showed that family members of OHS patients have normal chemosensitivity and respiratory centre behaviour, suggesting that reduced chemosensitivity is acquired.30 This may have therapeutic implications because treating OHS patients with NIV may reverse this impairment in chemosensitivity and improve blood gases in these patients.31–33
Although not all obese subjects will develop hypoventilation, a subgroup of morbidly obese subjects is susceptible to hypoventilation with hypercapnic respiratory failure, particularly when faced with new insults, such as a mild worsening of cardiac function or infection.5,34
PRESENTATION AND ASSESSMENT OF ARF IN OBESITY
The net result of the significant alterations in respiratory physiology in morbidly obese subjects is a diminished respiratory reserve that predisposes this group to ARF, even in the setting of minimal pulmonary or systemic insults. Obese subjects can present with hypoxaemic or hypercapnic ARF.
Acute hypoxaemic respiratory failure
Hypoxaemia is the most frequent gas exchange abnormality in obese subjects. It is more frequent in morbidly obese subjects and correlates well with the reduction in expiratory reserve volume.35 Obese subjects can develop acute hypoxaemic respiratory failure due to the common causes of respiratory distress, such as bronchial asthma, pulmonary oedema, pneumonia and postoperative complications, or as a part of progressive systemic diseases. The decreased oxygen reserves, and increased WOB and oxygen consumption in obesity predispose obese subjects to the development of hypoxaemic ARF following less severe and less extensive diseases compared with normal-weight individuals.18,36 Obesity is associated with more severe forms of asthma, and the response to standard treatment is reduced in obese patients compared with normal-weight asthmatics.37,38 Body mass index (BMI) has been reported as a determinant of asthma control independent of airway inflammation and obesity-related changes in lung mechanics.39 Furthermore, obesity has been associated with an increased risk of developing acute lung injury (ALI) and/or ARDS due to either pulmonary or extra-pulmonary causes.40,41 Several studies have reported that viral infections and pneumonia are more prevalent in obese subjects.42,43 In these cohorts, obese patients were susceptible to ARDS and severe multi-organ failure due to H1N1 infection.44,45 Hypoxaemic ARF in obesity can be mild and respond to supplemental oxygen therapy alone, or severe enough to warrant invasive mechanical ventilatory support. Recently, clinical and basic science researchers have begun to show interest in the relationship between obesity and ALI/ARDS. Because obesity and ALI/ARDS are both associated with alterations in inflammatory biomarkers, endothelial dysfunction and oxidative stress, investigators have raised the possibility that obese patients may be at higher risk of developing ALI/ARDS.46 ALI /ARDS in obese patients represents a new challenge to physicians because it requires a greater understanding of the mechanisms underlying lung injury in these patients to better characterize the syndrome and improve treatment. In a study assessing the impact of BMI on mechanically ventilated patients in the ICU, obesity (BMI 30–39.9 kg/m2) and severe obesity (BMI > 40 kg/m2) were significantly associated with the development of ARDS even after adjusting for confounders.40 This result is consistent with another report of increased hypoxaemic ARF and ARDS in obese patients after adjusting for other known risk factors, such as the ventilator settings.41 The pro-inflammatory markers released by adipocytes are thought to play an essential role in inducing ALI/ARDS.47 Aspiration, especially in the setting of difficult intubation, can contribute to the increased risk of developing ALI/ARDS.48 Nevertheless, obesity is generally not associated with increased mortality in ARDS.49
Obesity is a risk factor for acute and chronic hypercapnic respiratory failure.50,51 It is estimated that approximately 10% of obese subjects with a BMI between 30 and 40 kg/m2 have waking (daytime) hypercapnia.5 This proportion increases significantly in morbidly obese patients. It is difficult to predict which obese patients will develop hypercapnic respiratory failure.34,52 As in normal weight subjects, a number of respiratory disorders, such as COPD, chest wall abnormalities and neuromuscular disorders, can cause hypoventilation and present with acute hypercapnic respiratory failure in morbidly obese subjects. However, in this review, we focus on acute hypercapnic respiratory failure secondary to OHS. OHS is defined as the combined presence of obesity (BMI > 30 kg/m2) and arterial hypercapnia (PaCO2 > 45 mm Hg) while awake in the absence of other causes of hypoventilation.53 In addition to sleep hypoventilation, 90% of OHS patients have concurrent upper airway obstruction and OSA. However, a small minority of OHS patients have no evidence of OSA.54 The prevalence of OHS in the general population has not been determined. However, the prevalence of OHS among patients with OSA ranges from 10% to 20%, and in hospitalized patients with severe obesity, the prevalence reaches 31%.34,55 Typically, OHS patients present in the fifth or sixth decade of life with a history of dyspnoea on minimal exertion, hypersomnolence, and morning and nocturnal headaches.5 If the OHS patient also has OSA, the symptoms may include loud interrupted snoring, choking attacks during sleep and witnessed episodes of apnoea.5 Although OSA and OHS can coexist in the same patient, it is essential to distinguish OHS from pure OSA because treatment directed to eliminate airway obstruction in OSA may help some patients with OHS; however, a significant proportion of OHS patients may not improve with OSA treatment alone due to the persistence of nocturnal hypoventilation.56–59 In addition, portable unattended sleep studies and continuous positive airway pressure (CPAP) auto-titration trials are not recommended in OHS patients.60 Compared with BMI-matched eucapnic obese subjects, OHS patients have a higher incidence of respiratory and cardiovascular morbidity, a higher likelihood for hospitalisation, increased health-care utilisation and a significantly lower post-discharge survival rate.24,55,61
Because OHS is commonly under-recognized in routine practice, a significant proportion of OHS patients present at hospitals with ARF.31,55 Moreover, OHS is generally overlooked as a cause of hypercapnic ARF in morbidly obese patients; thus, many patients may be treated inappropriately without addressing hypoventilation and SBD.31,55,61,62
Table 1 presents a summary of studies of hypercapnic ARF in patients with OHS. Practitioners should consider OHS and SBD in all obese patients presenting with hypercapnic respiratory failure, and a detailed history from family members with regard to daytime sleepiness and witnessed apnoea should be obtained. There are no clearly defined clinical or laboratory predictors to determine which OHS patients are likely to develop hypercapnic ARF. However, the presence of daytime hypoxaemia and hypercapnia, significant oxygen desaturation during sleep, and comorbid conditions (e.g. significant pulmonary hypertension, cardiac function impairment and chronic lung disease) may increase the likelihood of developing acute hypercapnic respiratory failure.
Table 1. A summary of published studies on acute hypercapnic respiratory failure in patients with obesity hypoventilation syndrome (numbers represent mean values)
Other conditions that may cause hypoventilation that would lead to a different diagnosis should be excluded during the patient assessment, such as mechanical limitations (e.g. kyphoscoliosis or chronic lung disease), neurological conditions (e.g. neuropathies or diaphragmatic paralysis) or central control problems (e.g. hypothyroidism). Chest radiography is necessary to assess for chronic lung diseases. It generally shows reduced lung volumes and prominent pulmonary arteries when severe pulmonary hypertension is present. Obese patients with hypoxaemia should undergo arterial blood gas analysis to test for the presence of daytime hypercapnia. Assessment of arterial blood pH is essential for determining the severity of hypoventilation and guiding the level of care. Electrocardiography may reveal changes of pulmonary hypertension, and echocardiography is also useful for assessing left and right cardiac function, and pulmonary artery pressure. Laboratory testing should include a complete blood count to check for polycythemia, thyroid-stimulating hormone levels and serum electrolyte levels. Once the patient is stable, spirometry and respiratory muscle strength testing are essential to assess lung function to rule out respiratory causes of hypercapnia.
If acute ventilatory failure is suspected to be secondary to OHS, treatment should be commenced immediately without confirmation by a sleep study. However, a type I attended sleep study is essential once the patient is in stable condition to confirm the diagnosis and to prescribe the suitable positive airway pressure (PAP) device. Bi-level positive airway pressure (BPAP) is required in most OHS patients in the acute phase. However, many patients may benefit from CPAP once they are stable.69
MANAGEMENT OF ARF IN MORBIDLY OBESE PATIENTS
Proper management depends on an accurate initial assessment and diagnosis. The underlying disease process must be identified, and the proper medical therapy must be initiated. The compromised pulmonary function in obesity necessitates the provision of proper ventilatory support that will ensure adequate oxygenation and ventilation. In the next section, we discuss NIV ventilation in both hypoxaemic and hypercapnic (ventilatory) ARF, with emphasis on the set-up, monitoring and utilization of NIV in OHS patients with acute ventilatory decompensation. We then discuss some important issues related to invasive mechanical ventilation in obese patients, such as intubation, ventilatory strategies and weaning. Although there is no consensus for managing ARF in morbid obesity, we propose a systematic approach based on the available evidence and experiences.
NIV in the emergency management of ARF has been proven to be effective and safe in some respiratory and cardiac problems. It has been used successfully in acute cardiogenic pulmonary oedema resistant to standard medical therapy, even when accompanied by hypercapnia.70,71 It is also the standard of treatment for ARF secondary to acute exacerbations of COPD.72,73 The great advantage of NIV is the avoidance of endotracheal intubation and its associated complications, which are of particular concern in the obese, such as intubation- and extubation-related difficulties, ventilator-associated pneumonia, the need for sedative/muscle relaxants and difficult weaning.48,74–76 Therefore, in the absence of contraindications, NIV should be considered the first line of treatment for obese patients with ARF, particularly ventilatory failure. The anticipated outcome of NIV use depends on the type of ARF and the underlying cause. The two main PAP support modalities are CPAP and BPAP.
Acute hypoxaemic respiratory failure
Hypoxaemic ARF, regardless of the aetiology, is the most common clinical indication for endotracheal intubation. Frequently, patients with hypoxaemic ARF fail to respond to supplemental oxygen and require mechanical ventilation. Therefore, NIV has been used as an alternative option to endotracheal intubation, but the data are inconclusive. Data on hypoxaemic ARF in morbidly obese patients are scarce. Therefore, we will briefly present the available data on the utilization of NIV in hypoxemic ARF in general, and will then discuss in detail the acute ventilatory failure related to morbid obesity in OHS patients.
The available evidence indicates a reduced rate of endotracheal intubation with the use of NIV in ARF associated with COPD exacerbation, with cardiogenic pulmonary oedema and in immunocompromised patients presenting with fever and pulmonary infiltrates.77–79 In ARDS/ALI, the results are not very promising. In a multicentre, randomized, controlled trial of CPAP versus conventional oxygen therapy in patients presenting with hypoxaemic ARF (ALI, n = 102; cardiac diseases, n = 21), the early use of CPAP improved the physiological parameters (partial arterial oxygen concentration/FiO2) but did not reduce the endotracheal intubation rate, hospital mortality or ICU length of stay.80 Furthermore, CPAP use resulted in more adverse events before and after intubation, and several patients receiving CPAP therapy experienced cardiac arrest at the time of intubation.80 Overall, NIV in hypoxaemic ARF can be successful in selected patients, such as those with cardiogenic pulmonary oedema and some immunocompromised patients.77,79 In patients with severe illness, older age, ARDS or pneumonia, or those who failed to show any improvement in arterial blood gases after 1 h of treatment, the chance of NIV failure was high, and most of these patients required invasive ventilation.81
Acute hypercapnic respiratory failure
Hypercapnic ARF in the obese can be seen as a part of the clinical course of acute cardiogenic pulmonary oedema, pneumonia, bronchial asthma, COPD and other chronic respiratory diseases; however, in this section, we focus our discussion on the management of hypercapnic ARF complicating obesity hypoventilation.
Integration of the evidence regarding PAP, lung mechanics and the underlying pathological processes is essential to achieve successful results with PAP therapy. The general goals of treatment are to improve alveolar ventilation and upper airway patency, and to eliminate hypoxaemia. Oxygen therapy alone does not resolve hypoventilation and airway narrowing, and may worsen carbon dioxide retention, decrease minute ventilation, and increase apnoea and hypopnoea.31,67,82,83 In a recent double-blind, placebo-controlled, randomized crossover study in newly diagnosed OHS patients, Wijesinghe et al. demonstrated that the administration of 100% oxygen for 20 min increased transcutaneous carbon dioxide by 5 mm Hg and reduced minute ventilation by 13%.83
CPAP can be useful in mildly stable OHS patients as it may improve upper airway patency, increase FRC, offset auto-PEEP, unload respiratory muscles, reduce WOB and improve oxygenation through improving V/Q matching, but it usually does not resolve alveolar hypoventilation problems in OHS patients who have acute-on-chronic hypercapnic respiratory failure.13,31,57,59,67,69,84 BPAP therapy provides the same effects as CPAP therapy and also augments VT through the inspiratory component.
NIV set-up and monitoring.
Physicians should consider the possibility of OHS in all obese patients with hypercapnic ARF, and if OHS is suspected, NIV therapy should be initiated before performing a sleep study. The early initiation of NIV may avert the need for more invasive ventilation, thereby preventing its associated complications. The lack of an available ICU bed should not deter the start of NIV in this group of patients.85 If qualified staff and a monitoring set-up are available, NIV therapy can be started in the emergency room or the respiratory wards.86 Although new guidelines were recently published for stable chronic hypoventilation syndromes, there are no published consensus guidelines for hypercapnic ARF complicating obesity hypoventilation.87 Nevertheless, we will attempt to develop a systematic approach to the acute management of this under-recognized problem utilizing the best available evidence.
Careful mask fitting is essential to the success of an NIV trial. Nasal and oro-nasal masks have both been used effectively in the acute setting.67,88 However, patients with ARF usually prefer oro-nasal masks because these patients tend to be mouth breathers.89 Moreover, oro-nasal masks have the advantage of less air leakage and a higher VT, which result in better alveolar ventilation.90 Nevertheless, the interface should be tailored to the tolerance and preference of the patient.
Pressure- and volume-limited ventilations have been used in morbidly obese patients with hypercapnic ARF. However, based on the published data and our experience, we recommend starting with BPAP therapy. BPAP devices use a separately adjustable inspiratory (IPAP) and expiratory positive airway pressure (EPAP). The IPAP and EPAP levels are adjusted to maintain upper airway patency, and the pressure support (PS = IPAP-EPAP) augments ventilation.87 Both ICU ventilators and portable devices can be used in the acute setting; nevertheless, a device that can monitor VT, respiratory rate and air leakage is better able to assess ventilation.
Careful monitoring during PAP therapy, including vital signs, level of consciousness, respiratory pattern, oxygen saturation (using a pulse oximeter with a fast sampling rate (3–5 s)), delivered VT (if available) and arterial blood gases, is crucial to the success of therapy.
BPAP can be initiated with an EPAP of 4–6 cmH2O and an IPAP of 8–10 cmH2O. Then EPAP should be gradually increased until snoring, witnessed apnoeas and dips in oxygen saturation cease. Finally, IPAP can be gradually increased until an acceptable level of steady-state oxygen saturation (≥ 90%) is achieved.31,67 A maximum IPAP level of 30 cmH2O has been used in OHS patients with acute ventilatory failure.67,91 However, other studies have demonstrated that a maximum IPAP of 20 cmH2O is usually sufficient.31,69Figure 1 outlines the proposed management approach for OHS patients with acute ventilatory decompensation.
In general, 15–30% of patients with OHS require supplemental oxygen during sleep despite adequate PAP therapy.57,92 However, during acute respiratory decompensation, most OHS patients require supplemental oxygen.31,66,67 Therefore, if an OHS patient continues to have sustained hypoxaemia despite the delivery of high pressure to eliminate obstructive events and hypoventilation, oxygen should be provided through the mask to maintain oxygen saturation ≥ 90%. In the acute phase, NIV should be used continuously during the night and for 6–8 h during the day, allowing for verbal communication and the administration of medication and food. When daytime arterial blood gases are stable with an acceptable pH (≥7.35) and the patient's level of consciousness is acceptable, daytime NIV can be discontinued and replaced with low-flow supplemental oxygen.
The successful use of NIV in the context of acute ventilatory decompensation in OHS patients entails the following steps:
1Explain the indications and outcomes of NIV to the patient.
2Allow the patient to eat or drink nothing by mouth until he or she is stable to avoid the risk of aspiration.93
3Start by fitting the mask appropriately, and use low pressure for the initial acclimatisation to BPAP (IPAP 8–10 cmH2O, EPAP 4–6 cmH2O).
4Monitor heart rate, blood pressure, respiratory rate, VT (ideally >6–8 mL/kg) and oxygen saturation closely. Look for signs of respiratory distress.
5Increase EPAP gradually by 1–2 cmH2O to maintain a patent airway, and to prevent apnoea and dips in SaO2. IPAP should be increased concurrently by 1–2 cmH2O to maintain the PS > 4 cmH2O.
6If the patient continues to have persistent hypoxaemia (SaO2 < 90%), then IPAP should be increased to achieve SaO2 ≥ 90%. The PS (IPAP-EPAP) should be increased to 8–10 cmH2O to ensure proper ventilation, respiratory muscle rest and an adequate VT (6–8 mL/kg).
7Assess arterial blood gases 60 min after the initiation of NIV therapy. If the pH is increasing and PaCO2 is decreasing, the same setting should be continued. Transcutaneous CO2 monitoring could be utilized to monitor CO2 retention.94
8In patients who do not improve following this mode of ventilatory support, a trial of volume-cycled or average volume-assured pressure support ventilation can be attempted.58,95 If NIV fails, the patient should be considered for intubation and mechanical ventilation.
Volume-targeted pressure ventilation.
A limitation of BPAP is the inability to guarantee the delivered VT. Delivered VT could be compromised when fixed-level PS is used due to changes in respiratory system compliance in OHS patients during sleep and the possible variations with sleep position. To overcome this problem, a new hybrid mode of NIV that combines features of pressure and volume ventilation has been recently developed by several manufacturers to assure the delivery of a targeted VT to maintain good ventilation. The ventilator measures the delivered VT through a built-in pneumotachograph, and then automatically adjusts PS within a preset range to deliver VT close to the calculated targeted VT set by the treating physician. Different algorithms have been used; however, most ventilators utilize an algorithm that progressively adjusts the pressure level during several respiratory cycles to provide a VT close to the preset target volume.96 This new mode appears promising; however, the evidence supporting its effectiveness is still limited, particularly in patients with hypercapnic ARF. A randomized crossover study of 10 clinically stable OHS patients demonstrated that BPAP with volume targeting lowered PaCO2 compared with BPAP alone.58 However, there was no improvement in oxygenation, sleep quality or quality of life. An interim analysis of a randomized controlled trial of average volume-assured pressure support versus spontaneous-timed PS in OHS has been reported recently.97 Consecutive patients were randomized into spontaneous-timed or average volume-assured pressure support modes. Preliminary data for 43 patients (13 with hypercapnic ARF and 30 stable OHS) showed significant improvement in PaCO2 in average volume-assured pressure support patients, mean difference 0.63 kPa (P = 0.013); however the difference did not reach significance in the spontaneous-timed mode group (mean difference 0.55 kPa (P = 0.053)). A third randomized trial compared standard BPAP in an spontaneous-timed mode with volume targeted BPAP in a single night randomized crossover study in 12 stable OHS patients.98 Although volume targeting lowered nocturnal CO2, and increased average nocturnal VT and minute ventilation, there was a slight decrease in objective and subjective sleep quality, and comfort of ventilation.98 In our opinion, this new mode of ventilation can be tried in patients who continued to have hypoventilation despite adequate BPAP therapy. The suggested settings for average volume-assured pressure support include:
1Set the target VT to around 7–8 mL/kg of ideal body weight or 110% of the displayed VT when the patient is ventilated with spontaneous-timed mode, and then adjust settings according to the patient's response and tolerance.
2Set IPAP limits to a maximum of 30 cmH2O and a minimum of EPAP + 4 cmH2O.
3Set respiratory rate to achieve adequate ventilation. It is suggested to set a rate of 2–3 breaths per minute below resting respiratory rate.
4For control breaths, set inspiratory time between 30% and 40% if the patient has no COPD or asthma. If the patient has evidence of small airway obstruction, a shortened inspiratory time of 25–30% is suggested.
5Modify the above according to the patient's response and tolerance.
PAP therapy should be optimized later, once the patient is stable, based on type I attended sleep study findings and sleep specialist recommendations. In general, signs of improvement will appear within 1–2 h of initiating NIV therapy. Signs of improvement include improved mental status, oxygen saturation, PaCO2, and pH. Therefore, close monitoring for signs of improvement or deterioration is essential for patient safety. Major improvements in blood gas levels are usually noticed within a few days of initiating NIV therapy.
NIV may not be a suitable option for some obese patients with ventilatory failure, such as patients with haemodynamic instability, an unprotected upper airway, acute stroke or active upper gastrointestinal bleeding (Table 2). In addition, some patients may not tolerate NIV or may fail to respond to NIV.31,67 In such situations, endotracheal intubation and mechanical ventilation can be the only life-saving therapeutic modalities available.99 Intubating obese patients can be challenging and is associated with upper airway management difficulties. Furthermore, once the upper airway is secured, the proper ventilatory settings and modalities must be determined.
Table 2. Contraindications for non-invasive ventilation
• Altered mental status
• Orofacial deformities
• Haemodynamic instability
• Uncooperative patient
• Refractory hypoxaemia
• Upper gastrointestinal bleeding
• Inability to clear secretions
• Large acute myocardial infarction
• Uncontrolled arrhythmias
• Severe abdominal distension
• Acute stroke
Before the introduction of PAP therapy as a treatment option for SBD, tracheostomy was the main treatment modality for severe OSA and OHS. Tracheostomy is used to bypass the upper airway to mitigate the obstructive events.100 Tracheostomy is still considered a last treatment option for patients with severe OHS who cannot tolerate PAP therapy, have poor compliance with PAP therapy, or have failed weaning and extubation. However, it is important to realize that tracheostomy is not always effective in patients with SBD.56,100 In addition, tracheostomies in the obese must be undertaken with caution due to excess fat around the neck and the inferior larynx, and difficulty in identifying landmarks.7,100 Moreover, post-tracheostomy monitoring is essential to ensure resolution of the SBD.
Follow-up studies of OHS patients have demonstrated a significant improvement in arterial blood gases and lung mechanics after PAP therapy.31,67,101–103 Studies have shown that some OHS patients may not require daytime oxygen supplementation after several months of NIV therapy and that pulmonary artery pressure may decrease significantly in these patients.31,67,103 The precise physiological explanation for the improvement in respiratory parameters following adherence to PAP therapy is not clear, but several factors have been proposed, including the relief of upper airway obstruction64 as well as improvement in the following: lung volumes, particularly expiratory reserve volume and FRC;104 lung mechanics and V/Q matching by opening the atelectatic regions in the lung and altering the chest wall and lung mechanics;104 right ventricular function and pulmonary artery pressure via elimination of hypoxaemia;31,105 and central ventilatory control.106
Several studies have reported improved survival and a reduction in the need for hospitalization in treated OHS patients compared with untreated OHS patients.31,55,67,92,102,104 Nowbar et al. followed OHS patients for 18 months after discharge and demonstrated that mortality was 23% in the OHS group compared with 9% in the simple obesity group (hazard ratio = 4.0).55 Additionally, OHS patients were more likely to require ICU admission (40% vs 26%) and long-term care following discharge (19% vs 2%).55 Adherence to therapy is essential for a good outcome. Studies that followed OHS patients presenting with hypercapnic ARF after discharge found increased mortality in patients who refused PAP therapy.31,67 Perez de Llano et al., in a retrospective study of patients with OHS, demonstrated a 6% mortality rate in those who accepted and used PAP therapy (3 out of 54) compared with a 46% mortality rate in those who refused PAP therapy (7 out of 15).67 Long-term follow-up studies of OHS patients who were adherent to PAP therapy indicated that mortality was less than 10%.61,102,104 Recently, Priou and colleagues analyzed 4-year follow-up data on 92 stable OHS patients and 38 patients with acute exacerbation after ICU discharge.92 During the follow up, 24 patients (18.5%) discontinued NIV, and 24 patients (18.5%) died. Long-term compliance with NIV and oxygen therapy was lower in females, and female gender was the only independent predictor of mortality.92 No differences were found between stable OHS patients and OHS patients presenting with hypercapnic ARF in arterial blood gases, Epworth sleepiness scale score at 6 months, long-term survival and treatment adherence.92 This evidence suggests that long-term improvement is directly related to adherence to PAP treatment, which stresses the importance of close follow up and monitoring of adherence to PAP therapy in OHS patients.
Comorbid conditions involving other organs, such as the cardiovascular, respiratory and metabolic systems, are common among OHS patients and may influence the outcome. An improvement in haemoglobin levels and polycythaemia has been reported in OHS patients following PAP therapy.102,104
Invasive mechanical ventilation
One of the most challenging problems facing respirologists, intensivists and anaesthesiologists is the management of mechanical ventilation in obese ICU patients. Table 3 presents the potential respiratory problems of obese patients in the ICU. In general, obese ICU patients require mechanical ventilation more often than normal weight patients and for longer periods.76,108,109 However, some studies have shown no increase in mortality in ventilated obese ICU patients compared with normal-weight ICU patients.4,109
Table 3. Potential respiratory complications in morbidly obese patients in the intensive care unit
Prolonged duration of mechanical ventilation109,110
Airway management of obese patients
Morbid obesity and large neck circumference (>43 cm) have been identified as risk factors for difficult intubation,110 and this difficulty is significantly increased during emergency situations, when managing the airways of obese patients becomes very challenging. Obese patients cannot tolerate apnoea and tend to desaturate rapidly compared with normal-weight subjects due to reduced FRC, which results in low oxygen stores.111 Therefore, adequate pre-oxygenation becomes essential.112
The emergency room and the ICU should be properly equipped to manage the difficult airways of obese patients with ARF.112 Proper patient positioning before intubation is essential; studies have shown that head-up positioning improves oxygenation and lung function in morbidly obese patients. The ramped position has been proposed to improve oxygenation and the ability of the clinician to visualize the glottis, hence facilitating intubation in morbidly obese patients.112,113 The ramped position may be achieved by placing a blanket or a special device under the head and torso of the patient to horizontally align the external auditory meatus and the sternal notch.112
Ventilator management in obese patients with ARF
A discussion of mechanical ventilation in the morbidly obese is beyond the scope of this review. However, in this section, we attempt to stress the important practical points related to the mechanical ventilation of obese patients. While there are currently no standardized ventilatory protocols for obese patients, several studies have suggested ventilatory strategies for morbidly obese subjects. The majority of these studies have been performed in healthy obese patients undergoing bariatric surgery. Therefore, it is necessary to acknowledge that the ventilatory strategies applied in these subjects may not necessarily be effective in patients with lung diseases. An understanding of the lung mechanics associated with obesity is mandatory for the proper adjustment of mechanical ventilation settings, as discussed earlier.
Morbidly obese subjects have a low FRC, approaching the closing capacity of the small airways, which leads to collapse and atelectasis, particularly in dependent lung areas. Therefore, PEEP plays a vital role in recruiting respiratory units and reducing atelectasis in obese patients. One study showed that a PEEP of 10 cmH2O improved oxygenation and lung mechanics in morbidly obese subjects undergoing abdominal surgery, but not in normal-weight patients.114 Other investigators have since demonstrated a similar beneficial effect of PEEP.115–117 The effectiveness of higher levels of PEEP has also been reported.118 Lung recruitment manoeuvres using a transient increase in inspiratory pressure have been reported to reduce atelectasis and improve oxygenation when combined with PEEP in obese patients.115,119 In a recent randomized trial of 66 morbidly obese patients undergoing laparoscopy, Futier et al. demonstrated that a combination of pre-oxygenation with NIV pre-intubation and a lung recruitment manoeuvre of applying CPAP at 40 cmH2O for 40 s immediately after intubation resulted in improved lung volume and oxygenation during anaesthesia induction compared with pre-oxygenation alone or pre-oxygenation and NIV.120 There is a theoretical concern that creating high, positive intrathoracic pressure may impede the venous return to the heart and result in haemodynamic instability. However, intravascular volume expansion can be used to counteract this effect.121 Gernoth et al. have proposed that opening atelectatic regions in the lung may reduce pulmonary vascular resistance and increase cardiac output.122
There are no data to support the use of one mode of mechanical ventilation over another in obese subjects (e.g. volume-controlled vs pressure-controlled).9 However, if volume cycle ventilation is used in obese patients with ARF, a low VT is preferred, particularly in patients with ALI/ARDS. The protocol implemented in the ARDS clinical trial network (ARDS Network) recommended the use of a low VT (6 mL/kg predicted body weight) and the maintenance of a plateau pressure <30 cmH2O.123 These ventilator strategies were associated with an 8.8% absolute risk reduction in hospital mortality compared with conventional ventilation (VT = 12 mL/kg predicted body weight). A secondary analysis of the ARDS Network patients revealed that 58.6% of the studied population were overweight or had higher grades of obesity, and that the important primary and secondary outcomes were not significantly different between obese, overweight and normal-weight patients.124 The earlier data suggest that obese patients with ARF may be ventilated, as suggested by the ARDS Network Study protocol.123 It is important to remember that VT is calculated based on the ideal body weight and not the actual body weight because the size of the lung does not change much with increased weight.125 Another important consideration is estimation of the risk of ventilator-induced lung injury (barotrauma) in obese patients on mechanical ventilation. During positive pressure ventilation, two pressures are generated: i) transpulmonary pressure (airway pressure—pleural pressure), which is the distending pressure across the lung and is correlated with ventilator-induced lung injury126; and ii) transthoracic pressure (pleural pressure—atmospheric pressure), which is the distending pressure across the chest wall. In obese patients, the total compliance of the respiratory system is reduced due to reductions in both lung and chest wall compliance. Therefore, in obese patients, a marked increase in airway pressure may result in minimal lung distension and less risk of barotrauma due to increased chest wall elastance (stiffness), which causes a significant increase in pleural pressure, and thereby a decrease in transpulmonary pressure (airway pressure—pleural pressure).127 It is important to remember that the displayed, measured compliance reflects the compliance of the entire respiratory system and not only the lung compliance. Therefore, it may be difficult to stay within the recommended plateau pressure limits (<30 cmH2O) in obese patients. These changes may confer a lung protective effect, allowing for a higher PEEP and a higher plateau pressure to better ventilate obese patients.128
The position of obese patients during ventilation may affect lung mechanics and oxygenation.129 As discussed earlier, lung volumes and mechanics are compromised when obese subjects are in a supine position. Several studies have suggested that the ideal positioning for mechanically ventilated obese patients is the reverse Trendelenburg position, as long as there is no contraindication for this position. This position results in an increased VT, reduced respiratory rate and better oxygenation when compared with the supine and upright positions.130
Weaning from mechanical ventilation and extubation in obese patients are additional challenges for clinicians. Several factors may compromise weaning or extubation trials in obese patients. OSA and OHS are common among morbidly obese patients, which may lead to weaning failure, particularly in patients under the effects of sedatives or narcotics.131 In addition, weaning may be hampered by post-extubation hypoxaemia due to frequent basal atelectasis and high intra-abdominal pressure.132 Based on studies conducted in postoperative obese patients, we recommend the use of NIV post-extubation.133 El-Solh et al. demonstrated that the prophylactic use of NIV in severely obese patients in the first 48 h post-extubation in the ICU may be effective in avoiding respiratory failure.134 The reverse Trendelenburg position has also been reported to enhance extubation trial success.133,134
Obesity has a negative impact on the respiratory system and ventilatory control, which predisposes obese patients to an increased risk of developing acute respiratory decompensation. Obese subjects may present with hypoxaemic or hypercapnic respiratory failure. Data on the management of hypoxaemic ARF in obesity are limited. Moreover, data on the role of NIV in hypoxaemic respiratory failure are generally inconsistent. Acute hypercapnic respiratory decompensation secondary to obesity hypoventilation is an under-recognized disorder that responds well to NIV. The early recognition of this medical emergency and the initiation of NIV are essential in all obese patients who present with acute ventilatory failure to avoid the need for intubation and invasive mechanical ventilation. Good compliance with PAP therapy in obese patients with obesity hypoventilation improves lung function and blood gases, and may reduce the rates of hospital admission, morbidity and mortality in this population. The management of mechanical ventilation in morbidly obese patients is challenging; thus, understanding the pathophysiological changes of the respiratory system that occur in obesity is essential for successful ventilation and weaning of ventilated obese patients. PEEP and lung recruitment manoeuvres may counteract atelectasis in obese ventilated patients.
This work was partially supported by a grant from The National Program for Sciences, Technology, & Innovation (NPST), King Abdulaziz City for Science and Technology, and the University Sleep Disorders Center, King Saud University.