Clinical, functional and inflammatory characteristics in patients with paucigranulocytic stable asthma: Comparison with different sputum phenotypes

BACKGROUND

According to induced sputum cell count, four different asthma phenotypes have been recognized (eosinophilic, neutrophilic, mixed and paucigranulocytic). The aim of this study was to detect functional and inflammatory characteristics of patients with paucigranulocytic asthma.

METHODS

A total of 240 asthmatic patients were categorized into the four phenotypes according to cell counts in induced sputum. All patients underwent pulmonary function tests, and measurement of fraction of exhaled nitric oxide (FeNO). The levels of IL-8, IL-13 and eosinophilic cationic protein (ECP) were also measured in sputum supernatant. Treatment, asthma control and the presence of severe refractory asthma (SRA) were also recorded.

RESULTS

Patients were categorized into the four phenotypes as follows: eosinophilic (40%), mixed (6.7%), neutrophilic (5.4%) and paucigranulocytic (47.9%). Although asthma control test did not differ between groups (P=.288), patients with paucigranulocytic asthma had better lung function (FEV₁ % pred) [median (IQR): 71.5 (59.0-88.75) vs 69.0 (59.0-77.6) vs 68.0 (60.0-85.5) vs 80.5 (69.7-95.0), P=.009] for eosinophilic, mixed, neutrophilic and paucigranulocytic asthma, respectively, P=.009). SRA occurred more frequently in the eosinophilic and mixed phenotype (41.6% and 43.7%, respectively) and less frequently in the neutrophilic and paucigranulocytic phenotype (25% and 21.7%, respectively, P=.01). FeNO, ECP and IL-8 were all low in the paucigranulocytic, whereas as expected FeNO and ECP were higher in eosinophilic and mixed asthma, while IL-8 was higher in patients with neutrophilic and mixed asthma (P<.001 for all comparisons). Interestingly, 14.8% of patients with paucigranulocytic asthma had poor asthma control.

CONCLUSION

Paucigranulocytic asthma most likely represents a "benign" asthma phenotype, related to a good response to treatment, rather than a "true" phenotype of asthma. However, paucigranulocytic patients that remain not well controlled despite optimal treatment represent an asthmatic population that requires further study for potential novel targeted interventions.

Sputum interleukin-13 as a biomarker for the evaluation of asthma control

BACKGROUND

Asthma control refers to the extent to which the manifestations of asthma have been reduced or eradicated by treatment. Interleukin-13 (IL-13) has a central role in Th2 response and serves as a possible therapeutic target in uncontrolled asthma. Fraction of exhaled nitric oxide (FeNO) and sputum eosinophils have modest performance in the evaluation of asthma control.

OBJECTIVE

To assess the diagnostic performance of sputum IL-13 for the evaluation of asthma control and furthermore to investigate the performance of sputum eosinophils and FeNO.

METHODS

One hundred and seventy patients with asthma were studied. All subjects underwent assessment of asthma control by asthma control test (ACT), lung function tests, FeNO measurement and sputum induction for cell count identification and IL-13 measurement in supernatants.

RESULTS

IL-13 (pg/mL) levels in sputum supernatant differed significantly among patients with well-controlled asthma and those with not well-controlled asthma [median IQR 78 (66-102) vs. 213 (180-265), P < 0.001]. Receiver operating characteristic (ROC) analysis showed that, for the whole study population, the diagnostic performance of IL-13 was superior to both sputum eosinophils and FeNO levels [area under the curve (AUC) 0.92, 95% CI 0.87 to 0.95 vs. AUC 0.65, 95% CI 0.58 to 0.72 vs. AUC 0.65, 95% CI 0.55 to 0.72, respectively].

CONCLUSION

The diagnostic performance of sputum IL-13 was superior to both sputum eosinophils and FeNO levels for the identification of well-controlled asthma. Sputum IL-13 levels could serve as a useful biomarker for asthma control assessment.

Sputum and BAL Clara cell secretory protein and surfactant protein D levels in asthma

Clara cell secretory protein (CC16) is associated with Th2 modulation. Surfactant protein D (SPD) plays an important role in surfactant homeostasis and eosinophil chemotaxis. We measured CC16 and SPD in sputum supernatants of 84 asthmatic patients and 12 healthy controls. In 22 asthmatics, we additionally measured CC16 and SPD levels in BAL and assessed smooth muscle area (SMA), reticular basement membrane (RBM) thickness, and epithelial detachment (ED) in bronchial biopsies. Induced sputum CC16 and SPD were significantly higher in patients with severe asthma (SRA) compared to mild-moderate and healthy controls. BAL CC16 and SPD levels were also higher in SRA compared to mild-moderate asthma. CC16 BAL levels correlated with ED, while SPD BAL levels correlated with SMA and RBM. Severity represented a significant covariate for these associations. CC16 and SPD levels are upregulated in SRA and correlate with remodeling indices, suggesting a possible role of these biomarkers in the remodeling process.

Expiratory flow limitation in morbidly obese postoperative mechanically ventilated patients

Although obesity promotes tidal expiratory flow limitation (EFL), with concurrent dynamic hyperinflation (DH), intrinsic PEEP (PEEPi) and risk of low lung volume injury, the prevalence and magnitude of EFL, DH and PEEPi have not yet been studied in mechanically ventilated morbidly obese subjects. In 15 postoperative mechanically ventilated morbidly obese subjects, we assessed the prevalence of EFL [using the negative expiratory pressure (NEP) technique], PEEPi, DH, respiratory mechanics, arterial oxygenation and PEEPi inequality index as well as the levels of PEEP required to abolish EFL. In supine position at zero PEEP, 10 patients exhibited EFL with a significantly higher PEEPi and DH and a significantly lower PEEPi inequality index than found in the five non-EFL (NEFL) subjects. Impaired gas exchange was found in all cases without significant differences between the EFL and NEFL subjects. Application of 7.5 +/- 2.5 cm H2O of PEEP (range: 4-16) abolished EFL with a reduction of PEEPi and DH and an increase in FRC and the PEEPi inequality index but no significant effect on gas exchange. The present study indicates that: (a) on zero PEEP, EFL is present in most postoperative mechanically ventilated morbidly obese subjects; (b) EFL (and concurrent risk of low lung volume injury) is abolished with appropriate levels of PEEP; and (c) impaired gas exchange is common in these patients, probably mainly due to atelectasis.

Abdominal muscle fatigue after maximal ventilation in humans

Abdominal muscles are the principal muscles of active expiration. To investigate the possibility of abdominal muscle low-frequency fatigue after maximal ventilation in humans, we stimulated the nerve roots supplying the abdominal muscles. We used a magnetic stimulator (Magstim 200) powering a 90-mm circular coil and studied six normal subjects. To assess the optimum level of stimulation and posture, we stimulated at each intervertebral level between T7 and L1 in the prone, supine, and seated positions. At T10, we used increasing power outputs to assess the pressure-power relationship. Care was taken to avoid muscle potentiation. Twitch gastric pressure (Pga) was recorded with a balloon-tipped catheter. Mean (+/-SD) baseline twitch Pga measured with the subjects in the prone position at T10 was 23.5 +/- 5.4 cmH2O. Within-occasion mean twitch Pga coefficient of variation was 4.6 +/- 1.1%. Twitch Pga was measured with the subjects in the prone position with stimulation over T10 before and after 2 min of maximal isocapnic ventilation (MIV). Twenty minutes after MIV, mean twitch Pga fell by 17 +/- 9.1% (P = 0.03) and remained low 90 min after MIV. We conclude that after maximal ventilation in humans there is a reduction of twitch Pga and, therefore, of low-frequency fatigue in abdominal muscles.

Measurement of effective alveolar carbon dioxide tension during spontaneous breathing in normal subjects and patients with chronic airways obstruction

BACKGROUND

The measurement of effective alveolar carbon dioxide tension (PA-CO2eff) is still a matter of debate. It has, however, become common practice to use arterial instead of alveolar CO2 tension for computing alveolar oxygen tension (PAO2) and physiological dead space, not only in normal subjects but also in patients. The purpose of this study was to estimate alveolar CO2 tension during spontaneous breathing with a new bedside technique which is simple and non-invasive, and to compare these values with arterial CO2 tension measured in normal subjects and patients with chronic airways obstruction.

METHODS

The subjects breathed quietly through the equipment assembly (mouthpiece, monitoring ring, Fleisch transducer head) connected to a pneumotachograph and a fast response infrared CO2 analyser. The method is a computerised calculation of the volume weighted effective alveolar CO2 tension obtained from the simultaneously recorded expiratory flow and CO2 concentration versus time curves. An arterial blood sample was taken to measure PaCO2 for comparison during the study.

RESULTS

The results showed a mean difference (PACO2eff-PaCO2) of -0.205 kPa in 20 normal subjects and -0.460 kPa in 46 patients. The 95% confidence interval of the bias was -0.029 to -0.379 kPa in normal subjects and -0.213 to -0.707 kPa in patients. The limits of agreement between PACO2eff and PaCO2 were 0.526 to -0.935 in normal subjects and 1.170 to -2.088 in patients.

CONCLUSIONS

The volume weighted effective alveolar PCO2 in normal subjects and patients with chronic airways obstruction is lower than the arterial PCO2 and is recommended as a better estimate in the classical equations for estimating dead space and intrapulmonary shunt.

Detection of expiratory flow limitation during mechanical ventilation

Two new methods, application of negative pressure at the airway opening during expiration (NEP) and reduction of flow resistance by bypassing the expiratory line of the ventilator by exhaling into the atmosphere (ATM), were used to detect expiratory flow limitation in 12 semirecumbent (45 degree) mechanically ventilated patients, seven with chronic airway obstruction (CAO). An increase of expiratory flow with NEP or ATM, relative to the preceding control breath, was taken as indicating absence of expiratory flow limitation. By contrast, the portion of the tidal expiration over which there was no change in flow with NEP or ATM was considered as flow-limited. With NEP, nine patients exhibited flow limitation, six (all with CAO) were flow-limited over most of the tidal expiration (> 70% VT), and three at < 60% VT. Although the results with NEP and ATM were in general in good agreement, in the three non-flow-limited patients the ATM method gave erroneous results. Six patients were also studied supine, including two who were not flow-limited when semirecumbent: both became flow-limited when supine. We conclude that NEP provides a simple method to detect flow limitation in mechanically ventilated patients. The supine position enhances flow limitation.

Lack of effect of acute hypoxia and hypercapnia on muscle relaxation rate in man

The effect of acute hypoxia and hypercapnia on skeletal muscle relaxation rate was investigated in man. The contractile force of limb muscles (quadriceps femoris and adductor pollicis) was studied in normal subjects using electrical stimulation. The twitch transdiaphragmatic pressure was used to investigate the diaphragm. The maximum relaxation rate was measured from contractions of the non-fatigued quadriceps femoris, adductor pollicis and diaphragm. Severe hypoxia (mean transcutaneous oxygen tension 40 mmHg) did not alter the maximum relaxation rate from 20 Hz stimulated contractions of the non-fatigued quadriceps. Acute hypercapnia (mean end-tidal CO2 tension 65.1 mmHg) did not significantly change the maximum relaxation rate from 20 and 100 Hz tetanic contractions of non-fatigued quadriceps. The maximum relaxation rate from 1 Hz twitch tension and 20 Hz tetanic tension of adductor pollicis was also not altered by acute hypercapnia (mean end-tidal carbon dioxide tension 63 and 60 mmHg, respectively). The maximum relaxation rate from twitch transdiaphragmatic pressure was not altered by acute hypercapnia (mean end-tidal carbon dioxide tension 69.7 mmHg).

Effect of acute hypercapnia on limb muscle contractility in humans

The effect of acute hypercapnia on skeletal muscle contractility and relaxation rate was investigated. The contractile force of fresh and fatigued quadriceps femoris (QF) and adductor pollicis (AP) was studied in normal humans by use of electrical stimulation. Maximum relaxation rate from stimulated contractions was measured for both muscles. Acute hypercapnia led to a rapid substantial reduction of contraction force. The respiratory acidosis after 9% CO2 was breathed for 20 min [mean venous blood pH 7.26 and end-tidal PCO2 (PETCO2) 65.1 Torr] reduced 20- and 100-Hz stimulated contractions of QF to 72.8 +/- 4.4 and 80.0 +/- 5.1% of control values, respectively. After 8 and 9% CO2 were breathed for 12 min, AP forces at 20- and 50-Hz stimulation were also reduced. Twitch tension of AP was reduced by a mean of 25.5% when subjects breathed 9% CO2 for 12 min [mean arterialized venous blood pH (pHav) 7.25 and PETCO2 66 Torr]. Over the range of 5% (pHav 7.38 and PETCO2 47 Torr) to 9% CO2, there was a linear relationship between twitch tension loss and pHav, arterialized venous blood PCO2, and PETCO2. Acute respiratory acidosis (mean PETCO2 61 Torr) increased the severity of low-frequency fatigue after intermittent voluntary contractions of AP. At 20 min of recovery, twitch tension was 63.2 +/- 13.4 and 46.8 +/- 16.4% of control value after exercise breathing air and 8% CO2, respectively. Acute hypercapnia (mean PETCO2 65.1 and 60.5 Torr) did not alter the maximum relaxation rate from tetanic contractions of fresh QF and from twitch tensions of AP.

The effect of posture and abdominal binding on respiratory pressures

We examined the effect of posture on the generation of respiratory pressures in 6 highly trained subjects. Transdiaphragmatic pressure was measured at FRC during bilateral percutaneous phrenic nerve stimulation (twitch Pdi) and maximal sniffs (sniff Pdi), with the abdomen bound and unbound. Maximum static inspiratory (PImax) and expiratory (PEmax) mouth pressures were measured with the abdomen unbound. Three postures were examined: seated (Se), semi-supine (30s), and supine (Su). Changes of posture did not significantly alter twitch Pdi. By contrast, sniff Pdi and static mouth pressures were significantly reduced in the Su posture. Abdominal binding significantly increased twitch Pdi only. We conclude that voluntary respiratory manoeuvres requiring activation, recruitment and coordination of different muscle groups are performed better in the Se position. We suggest that posture be standardised for serial comparative measurements of voluntary respiratory pressures in a given subject.

The effect of posture and abdominal binding on respiratory pressures

We examined the effect of posture on the generation of respiratory pressures in 6 highly trained subjects. Transdiaphragmatic pressure was measured at FRC during bilateral percutaneous phrenic nerve stimulation (twitch Pdi) and maximal sniffs (sniff Pdi), with the abdomen bound and unbound. Maximum static inspiratory (PImax) and expiratory (PEmax) mouth pressures were measured with the abdomen unbound. Three postures were examined: seated (Se), semi-supine (30s), and supine (Su). Changes of posture did not significantly alter twitch Pdi. By contrast, sniff Pdi and static mouth pressures were significantly reduced in the Su posture. Abdominal binding significantly increased twitch Pdi only. We conclude that voluntary respiratory manoeuvres requiring activation, recruitment and coordination of different muscle groups are performed better in the Se position. We suggest that posture be standardised for serial comparative measurements of voluntary respiratory pressures in a given subject.

Maximal relaxation rates of esophageal, nose, and mouth pressures during a sniff reflect inspiratory muscle fatigue

Maximal relaxation rate (MRR, % pressure fall/10 msec) of the inspiratory muscles is reduced with fatigue. We have investigated whether MRR of esophageal pressure (Pes) generated by voluntary sniffs is decreased by fatigue, and whether sniff nasopharyngeal (Pnp) and mouth (Pmo) MRR reflect these changes. In 10 normal subjects, control MRR of sniff Pes correlated closely to Pnp MRR (r = 0.977, p less than 0.001) and Pmo MRR (r = 0.947, p less than 0.001). To produce inspiratory muscle fatigue, four highly motivated subjects breathed to exhaustion (3 to 6 min) through a high inspiratory resistance. MRR was determined from 10 sniffs for Pes, Pnp, and Pmo before fatigue, and at intervals up to 10 min after fatigue. The subjects showed a mean decrease in sniff Pes MRR of 33% (range, 20 to 42) immediately after fatigue, which returned exponentially to control values within 3 to 4 min. The mean changes in Pes MRR were reflected by similar changes in Pnp MRR, 32% (range, 18 to 43) and Pmo MRR, 33% (range, 21 to 42). Studies were repeated in the four subjects with closely similar results. We conclude that fatigue of the inspiratory muscles reduces MRR of sniff Pes, and that this is reflected in Pnp and Pmo. Sniff Pes, Pnp, and Pmo MRR measurements may provide a useful method for detecting and monitoring fatigue; Pnp and Pmo have the advantage of being less invasive.

The measurement of inspiratory muscle strength by sniff esophageal, nasopharyngeal, and mouth pressures

Sniff esophageal pressure (Pes) is a useful measurement of global inspiratory muscle strength, although it does require passage of an esophageal balloon. We investigated the relationship between nasopharyngeal pressure (Pnp) or pressure within the mouth (Pmo) and Pes during a maximal sniff from FRC without a noseclip. We measured Pes, Pnp, and Pmo simultaneously in 10 normal volunteers, and in 12 patients with inspiratory muscle weakness. In both groups, Pnp and Pmo were slightly less but very close to Pes. In normal volunteers, the mean ratio Pnp/Pes was 0.92 +/- 0.006 (mean +/- SE) and Pmo/Pes was 0.95 +/- 0.006. Regression analysis showed Pes = 4.57 + 1.05 Pnp (r = 0.995, p less than 0.001) and Pes = 0.74 + 1.05 Pmo (r = 0.994, p less than 0.001). Similar relationships between Pnp, Pmo, and Pes were found over a wide range of pressures generated by submaximal sniffs in normal subjects. In patients, the mean ratio Pnp/Pes was 0.90 +/- 0.02 and Pmo/Pes was 0.87 +/- 0.03. Regression analysis showed Pes = 5.12 + 1.0 Pnp (r = 0.949, p less than 0.001) and Pes = 11.2 + 0.882 Pmo (r = 0.936, p less than 0.001). We conclude that Pnp and Pmo predict Pes during a maximal sniff in both normal subjects and in patients with inspiratory muscle weakness. Sniff Pnp and/or Pmo may provide a useful and less invasive method of measuring maximal inspiratory pressures during a sniff.

Comparison of two different mouthpieces for the measurement of Pimax and Pemax in normal and weak subjects

We investigated the effect of mouthpiece design on maximum static expiratory (PEmax) and inspiratory (PImax) mouth pressures. We measured PEmax from total lung capacity (TLC) and PImax from residual volume (RV) in 21 healthy volunteers, and in 40 patients referred for respiratory muscle testing. We compared two different mouthpieces, a semi-rigid plastic flanged type fitting inside the lips, and a 4 cm diameter rubber tube held against the lips. The tube mouthpiece gave significantly higher values for PEmax (p less than 0.02) in all subjects. PImax was also significantly higher (p less than 0.005) with the tube mouthpiece in subjects who recorded normal pressures. We conclude that maximum pressures are obtained in all normal subjects with the rubber tube mouthpiece, and that differences in quoted normal ranges of maximum static respiratory pressures reflect in part the design of the mouthpiece and the way in which it was used.

Comparison of two different mouthpieces for the measurement of Pimax and Pemax in normal and weak subjects

We investigated the effect of mouthpiece design on maximum static expiratory (PEmax) and inspiratory (PImax) mouth pressures. We measured PEmax from total lung capacity (TLC) and PImax from residual volume (RV) in 21 healthy volunteers, and in 40 patients referred for respiratory muscle testing. We compared two different mouthpieces, a semi-rigid plastic flanged type fitting inside the lips, and a 4 cm diameter rubber tube held against the lips. The tube mouthpiece gave significantly higher values for PEmax (p less than 0.02) in all subjects. PImax was also significantly higher (p less than 0.005) with the tube mouthpiece in subjects who recorded normal pressures. We conclude that maximum pressures are obtained in all normal subjects with the rubber tube mouthpiece, and that differences in quoted normal ranges of maximum static respiratory pressures reflect in part the design of the mouthpiece and the way in which it was used.