Which clinical signs and symptoms predict hypoxemia in acute childhood asthma
http://www.100md.com
《美国医学杂志》
1 Department of General Practice, Maastricht University, The Netherlands,2 Post Graduate Institute of Medical Research and Education (PGIMER), Department of Pediatrics, Chandigarh, India,
Abstract
Objective. To find the clinical signs that are the best predictors of hypoxemia (SpO[2] =92%) in acute asthma in children. Methods. Color of skin, dyspnea (by single breath counting), alertness, respiratory rate, presence of audible wheeze, wheezing on auscultation, accessory muscle use, nasal flaring, pulse rate, systolic and diastolic blood pressure, pulsus paradoxus and oxygen saturation at room air (by pulse oximetry) were recorded at the time of presentation and one hour after presentation after completion of 3 doses of nebulized salbutamol and budesonide. Results. Hypoxemia (SpO[2] ￡ 92%) was seen in 45% children at presentation and 14(28.6%) after one hour. The clinical signs that correlated significantly with hypoxemia at both time points were dyspnea assessed by single breath count (OR 3.3, 95% CI 0.9-12.9), accessory muscle use score 3 3 (OR 3.0, 95% CI 0.9-15.4) and pulsus paradoxus >10 (OR 3.0, 95% CI 0.7-13.6). In a multiple logistic regression model accessory muscle score 3 3 and pulsus paradoxus >10 were identified as independent predictors of hypoxemia (sensitivity 64.3%, specificity 91%). Conclusion. Physical assessment in a child with acute exacerbation of asthma should at least include accessory muscle use and pulsus paradoxus, since these predict hypoxemia the best.
Keywords: Acute asthma; Oxygen saturation; Pulsus paradoxus; Hypoxemia; Pediatric emergency
Asthma is the most common chronic disease in children.[1] The management of acute exacerbation of asthma rely on the careful ongoing clinical assessment, complemented by serial measurement of lung function.[2] The danger of inadequate assessment of acute asthma is that more effective therapy may be withheld in the acute stage, with possible increase in hospitalization and mortality rates. Since medical history taking and physical examination are only moderately effective in diagnosing and estimating severity of acute asthma, objective measures of lung function are necessary for the accurate assessment.[3] Clinically the severity of an acute exacerbation of asthma is assessed by presence of tachypnea, tachycardia, a reduction of more than 40% in expected peak expiratory flow rate (PEFR), or an inspiratory fall of arterial blood pressure of >10 mmHg (pulsus paradoxus).[4] More recently the value of oxygen saturation by pulse oximetry (SpO 2 ) in the assessment of acute severe asthma has been a subject of study,[5],[6] it is said to be a good predictor of prolonged bronchodilator therapy and hospitalization.[7],[8],[9] Since oxygen saturation (SpO 2 ) and sometimes even PEFR measurement may not be available to physicians/health care provider in developing countries especially in primary health care setting, they have to rely on clinical assessment based on simple physical signs at presentation. These physical signs include cough, the color of the skin (cyanosis), nasal flaring, degree of dyspnea, alertness level, fragmentation of speech, respiratory rate, accessory muscle use, amount of both audible and auscultational wheeze and heart rate. However, these signs have not been fully studied for their ability to predict hypoxemia and severity of the asthma attack. The objective of this study was to find which of the clinical signs mentioned above are the best predictors of hypoxemia (defined as SpO[2] < 92%) in an acute asthma attack.
Materials and Methods
Population: Children between 3 to 15 years of age, with acute moderate or severe exacerbation of bronchial asthma, presenting to the pediatric emergency service of a tertiary care teaching hospital in north India from April to June 2003 were included in the study. The study was approved by Institute's Ethics Committee. Informed consent was obtained from the parents. All children were treated according to a standard protocol using salbutamol nebulization (0.15 mg/Kg/dose with a minimum of 2.5 mg) and budesonide (800 μg) nebulization along with oxygen 5 L/min given through a simple facemask. This treatment was repeated every 20 minutes for 3 doses. In case of no or inadequate improvement nebulized ipratropium bromide (250 mg every 20 min) and hydrocortisone (10 mg/kg, I.V.) were added. A detailed history including the complaints of the patients and medication use were recorded. The children were assessed at the time of presentation (0 hour) and at 1 hour after presentation (after 3 doses of nebulized salbutamol + budesonide).
Measurements : The color of the skin was assessed by observing the lips and tongue (pink, blue or pale). Dyspnea was assessed by the ability to count from one to ten slowly in one deep breath (maximum number count was recorded) and the ability to speak in full sentences. Alertness was noted as normal, when the child was able to answer questions and interact, and obtunded when it was difficult to get the child to answer or interact with the examiners. Respiratory rate was counted for one full minute, by looking at the child's abdomen and counting the numbers of each upward movement as one breath. Auscultational wheez was assessed by stethoscope and for its presence during expiration and inspiration, and location. It was graded with help of an ordinal score.[10] Accessory muscle use was noted by observation of supra-clavicular, inter-costal and sub-costal retractions[11]. Each of these items were divided into absent, mild, moderate and marked[10]. Nasal flaring was noted as either absent or present assessed by observing movement of the nostrils for 30 sec. Pulse rate was counted by palpation and heart-rate per min by the pulse oximeter (Nonin 8500). For blood pressure measurement and pulsus paradoxus mercury sphygnomano-meters with small cuffs (10 cm) were used for the younger children and normal cuffs (15 cm) for the older children. Pulsus paradoxus was defined as a decrease in blood pressure of more than 10 mmHg during inspiration. The SpO2 was measured while the child was breathing room air, with a pulse oximeter (Nonin 8500) using a finger clip on the second finger of the right hand. All the observation about clinical signs were made and recorded by MSR and RPG who were blind to SpO 2 measurement, which was measured by a research staff. In 73% of the cases there was no significant inter-observer difference in various clinical signs measured; in the remaining cases the measurement was repeated and the mean of the two values measured was recorded.
For analysis most symptoms were categorized into a dichotomus scale with values of 0 and 1 table1. Auscultational wheeze was recorded objectively using a scoring system in which children could score between 0-8 points.[10] Wheeze during expiration was categorized as '0' for no wheeze, 1 for end expiratory wheeze, 2 for wheeze during half of the expiration, 3 for wheeze during three-fourth of the expiration and 4 for wheeze during the entire expiration. Likewise, wheezing during inspiration was categorized into 3 categories '0' for no wheeze during inspiration, 1 for wheeze in a part of inspiration and 2 for wheeze during the entire inspiration. Finally, the location of the wheezing sound in the lung was categorized in '0' for no wheeze, 1 for wheeze in 1 or 2 segments of the lung and 2 for wheeze audible in 3 or 4 segments of the lung. By adding up the individual scores of each item, a total wheeze score was acquired. For accessory muscle a child could score 0-9 points[10]. Supra-clavicular, intercostals and sub-costal retractions were categorized into, '0' for none, 1 for mild, 2 for moderate and 3 for marked retractions. By adding up the individual scores of supra-clavicular, intercostals and sub-costal retractions, a total accessory muscle use score was acquired.
Analysis : All the gathered data were analyzed at two time points: at presentation and after one hour. Hypoxemia was defined as a SpO 2 ￡ 92%. Descriptive statistics (mean ± SD) and percentage was used for presentation of data; t-test was used for 'between the groups' comparison of continuous variables and x 2sub -test for discreet variables. The associations between various clinical signs and hypoxemia was evaluated by calculating odds ratios (OR) and 95 percent confidence intervals (95% CI). For this purpose tachypnoea was defined as a respiratory rate > 40/min in children 1 year to 5 years, and > 30/ min in older children. Ordinal scores on auscultational wheezing and accessory muscle use were changed to discrete variable using a cut off, which reflected mild and severe grade of the sign. Cut off score for wheezing was <4 and 3 4, and for accessory muscle use was <3 and 3 3. The variable that were significant on univariate analysis were entered in a multiple logistic regression analysis to identify independent predictors of hypoxemia. All analysis were done using SPSS version 11 and Epi-info 2000.
Results
A total of 51 consecutive patients (34 boys and 17 girls) were assessed at two time points, at admission and after completion of three dosage of salbutamol and budesonide nebulization. The patients were between 29 and 186 months old. (median 82 months and mean ± SD 90.5 ± 38.8 months). The clinical profile of patients at presentation and after one hour is given in table2.
Hypoxemia was seen in 23 (45%) patients at admission and 14 (28.6%) patients after 3 dosage of nebulization. Single breath count, accessory muscle use, dyspnoea, auscultatory wheezing score 34 and pulsus pradoxus were the clinical signs that showed significant association with hypoxemia at presentation and after one hour table3. All the above variables except single breath counting were entered in a logistic regression analysis (backward stepwise) for both the time points using hypoxemia as dependent variable. At both the time points accessory muscle use (score 3 3) and pulsus paradoxus were identified as the two independent predictors of hypoxemia table4. None of the model was highly predictive of hypoxemia table4. Alertness, respiratory rate, auscultational wheezing, cough, pulse, systolic and diastolic blood pressure, and audible wheezing, did not correlate with hypoxemia.
Discussion
Our study shows that at presentation and after initial bronchodilator therapy pulsus paradoxus and accessory muscle use were significantly related to the hypoxemia. Other clinical signs namely single breath count, auscultatory wheeze score 3 4 and fast pulse rate also did reach significance individually but on multiple logistic regression these factors did not stand out as independent predictors of hypoxemia in this population. Single breath count (a reflection of dyspnea) may have reached significance in a larger population sample but it was available in only 50% of the study populations; at younger ages it was not possible to assess this sign properly. Our finding is important since both signs are easy to measure in primary care and may be used as clinical guideposts in the initial assessment of acute childhood asthma. Although both variables have been mentioned in literature as marker of severity of acute asthma, this is the first time that an evidence base is provided in comparison with other clinical variables for evaluation of hypoxemia in acute asthma.
Previous studies on the predictive value of pulsus paradoxus in asthma showed contradictory results.[12],[13],[14],[15] In our study pulsus paradoxus showed a highly significant association with oxygen saturation both at presentation and after 1 hour. It may be considered as a very valuable clinical sign, both in assessing the severity of an asthma attack and predicting hypoxemia.
Impaired alertness and audible wheeze were not common and did not correlate with oxygen saturation at any time. With the exception of one case, none of the patients showed impaired alertness, even when they were hypoxemic. Therefore, alertness and audible wheeze cannot be considered as useful signs in assessing the severity of acute asthma in children. In the way that we categorized the continuous values of respiratory rate, pulse, systolic and diastolic blood pressure and auscultational wheeze none of these symptoms showed any significant correlations with oxygen saturation.
There are some shortcomings in our study. First, we expected to include more patients in our study. Patients came in at a rate of 2-4 per day in April and the inflow decreased to 0-1 per day in June. Nevertheless we believe that the number of patients studied, was sufficient for measuring the association and coming to conclusions. Another possible consequence of the seasonal manner of patient inflow is that there might be a different etiology of asthma in these patients. However, this is considered less relevant to our study, since we were studying the predictability of oxygen saturation by the clinical symptoms in children with asthma, irrespective of the etiology. Our study was conducted in a hospital, which serves both as a referral centre as well as a first contact health care facility. This might result in some selection bias: patients with more severe attacks might have been referred to us. However, since the aim of our study was to assess the symptoms in an acute attack, we did not consider this as a serious limitation to our conclusion and outcome.
We conclude that in assessing an asthma exacerbation, a full clinical assessment of the patient including oximetry is warranted. However, in the absence of a pulse oximeter or peak flow, pulsus paradoxus and accessory muscle use (score 3 3) seems to deserve priority in assessing presence of hypoxemia. We recommend further studies to assess our results in a rural, primary care setting, using a larger sample.
Acknowledgments
We are grateful to the staff of emergency department of PGIMER for their hospitality during our stay in Chandigarh and their help and support during the execution of our study. Special thanks to Dr. Herpreet Kaur Arora and all the patients who participated
References
1. The International Study of Asthma and Allergies in Childhood (ISAAC); Worldwide Variation in Prevalence of symptoms of asthma, allergic rhinoconjuctivitis, and atopic eczema. Lancet 1998; Apr: 1225-1232.
2. Canny GJ, Reisman J, Healy R et al. Acute Asthma: Observations regarding the management in a Pediatric Emergency Room. Paediatrics 1989; 83: 507-512.
3. Li JT, O'Connell EJ. Clinical evaluation of asthma. Pediatr Child Health 1995; 31 : 131-133.
4. Hetzel M, Modell M. Asthma. Occas Pap R Coll Gen Pract 1992; 58 : 14-20.
5. Wright RO, Santucci KA, Jay GD et al. Evaluation of pre- and post-treatment pulse oximetry in acute childhood asthma. Acad Emerg Med 1997; 4 : 114-117.
6. Sole D, Komatsu MK, Carvalho KV, Naspitz CK. Pulse oximetry in the evaluation of the severity of acute asthma and/or wheezing in children. J Asthma 1999; 36 : 327-333.
7. Mehta SV, Parkin PC, Stephens D, Schuh S. Oxygen saturation as a predictor of prolonged, frequent bronchodilator therapy in children with acute asthma. J Pediatr 2004; 145 : 641-645.
8. Keogh KA, Macarthur C, Parkin PC et al. Predictors of hospitalization in children with acute asthma. J Pediatr 2001; 139 : 273-277.
9. Keahey L, Bulloch B, Becker AB et al. Multicenter Asthma Research Collaboration (MARC) Investigators. Initial oxygen saturation as a predictor of admission in children presenting to the emergency department with acute asthma. Ann Emerg Med 2002; 40 : 300-307.
10. Lowell DI, Lister G, Von Kross H et al. Wheezing in infants: the response to epinephrine. Pediatrics 1987; 79 : 939-945.
11. Sethi GR, Choudhary P, Sachdev HPS. Chapter 7- Acute severe asthma. In: Sachdev HPS, Puri RK, Bagga A edrs. Principles of Pediatric and Neonatal Emergencies . Jaypee Brothers. New Delhi, 1994, pp.71-84.
12. Levi M, Hart W, Wieling W. Physical diagnosis - paradoxical pulse. Ned Tijdschr Geneesk 1999; 143 : 2045-2048.
13. Frey B, Freezer N. Diagnostic value and pathophysiologic basis of pulsus paradoxus in infants and children with respiratory disease. Pediatr Pulmonol 2001; 31 : 138-143.
14. Wright RO, Steele DW, Santucci KA et al. Continuous, non-invasive measurement of pulsus paradoxus in patients with acute Asthma. Allergy Asthma Immunol 1996; 76 : 1-13.
15. Chavannes N. Pulse oximetry and respiratory disease in primary care. Prim Care Resp J 2003; 12 : 2-3.(Rahnama'i MS, Geilen RP, )
Abstract
Objective. To find the clinical signs that are the best predictors of hypoxemia (SpO[2] =92%) in acute asthma in children. Methods. Color of skin, dyspnea (by single breath counting), alertness, respiratory rate, presence of audible wheeze, wheezing on auscultation, accessory muscle use, nasal flaring, pulse rate, systolic and diastolic blood pressure, pulsus paradoxus and oxygen saturation at room air (by pulse oximetry) were recorded at the time of presentation and one hour after presentation after completion of 3 doses of nebulized salbutamol and budesonide. Results. Hypoxemia (SpO[2] ￡ 92%) was seen in 45% children at presentation and 14(28.6%) after one hour. The clinical signs that correlated significantly with hypoxemia at both time points were dyspnea assessed by single breath count (OR 3.3, 95% CI 0.9-12.9), accessory muscle use score 3 3 (OR 3.0, 95% CI 0.9-15.4) and pulsus paradoxus >10 (OR 3.0, 95% CI 0.7-13.6). In a multiple logistic regression model accessory muscle score 3 3 and pulsus paradoxus >10 were identified as independent predictors of hypoxemia (sensitivity 64.3%, specificity 91%). Conclusion. Physical assessment in a child with acute exacerbation of asthma should at least include accessory muscle use and pulsus paradoxus, since these predict hypoxemia the best.
Keywords: Acute asthma; Oxygen saturation; Pulsus paradoxus; Hypoxemia; Pediatric emergency
Asthma is the most common chronic disease in children.[1] The management of acute exacerbation of asthma rely on the careful ongoing clinical assessment, complemented by serial measurement of lung function.[2] The danger of inadequate assessment of acute asthma is that more effective therapy may be withheld in the acute stage, with possible increase in hospitalization and mortality rates. Since medical history taking and physical examination are only moderately effective in diagnosing and estimating severity of acute asthma, objective measures of lung function are necessary for the accurate assessment.[3] Clinically the severity of an acute exacerbation of asthma is assessed by presence of tachypnea, tachycardia, a reduction of more than 40% in expected peak expiratory flow rate (PEFR), or an inspiratory fall of arterial blood pressure of >10 mmHg (pulsus paradoxus).[4] More recently the value of oxygen saturation by pulse oximetry (SpO 2 ) in the assessment of acute severe asthma has been a subject of study,[5],[6] it is said to be a good predictor of prolonged bronchodilator therapy and hospitalization.[7],[8],[9] Since oxygen saturation (SpO 2 ) and sometimes even PEFR measurement may not be available to physicians/health care provider in developing countries especially in primary health care setting, they have to rely on clinical assessment based on simple physical signs at presentation. These physical signs include cough, the color of the skin (cyanosis), nasal flaring, degree of dyspnea, alertness level, fragmentation of speech, respiratory rate, accessory muscle use, amount of both audible and auscultational wheeze and heart rate. However, these signs have not been fully studied for their ability to predict hypoxemia and severity of the asthma attack. The objective of this study was to find which of the clinical signs mentioned above are the best predictors of hypoxemia (defined as SpO[2] < 92%) in an acute asthma attack.
Materials and Methods
Population: Children between 3 to 15 years of age, with acute moderate or severe exacerbation of bronchial asthma, presenting to the pediatric emergency service of a tertiary care teaching hospital in north India from April to June 2003 were included in the study. The study was approved by Institute's Ethics Committee. Informed consent was obtained from the parents. All children were treated according to a standard protocol using salbutamol nebulization (0.15 mg/Kg/dose with a minimum of 2.5 mg) and budesonide (800 μg) nebulization along with oxygen 5 L/min given through a simple facemask. This treatment was repeated every 20 minutes for 3 doses. In case of no or inadequate improvement nebulized ipratropium bromide (250 mg every 20 min) and hydrocortisone (10 mg/kg, I.V.) were added. A detailed history including the complaints of the patients and medication use were recorded. The children were assessed at the time of presentation (0 hour) and at 1 hour after presentation (after 3 doses of nebulized salbutamol + budesonide).
Measurements : The color of the skin was assessed by observing the lips and tongue (pink, blue or pale). Dyspnea was assessed by the ability to count from one to ten slowly in one deep breath (maximum number count was recorded) and the ability to speak in full sentences. Alertness was noted as normal, when the child was able to answer questions and interact, and obtunded when it was difficult to get the child to answer or interact with the examiners. Respiratory rate was counted for one full minute, by looking at the child's abdomen and counting the numbers of each upward movement as one breath. Auscultational wheez was assessed by stethoscope and for its presence during expiration and inspiration, and location. It was graded with help of an ordinal score.[10] Accessory muscle use was noted by observation of supra-clavicular, inter-costal and sub-costal retractions[11]. Each of these items were divided into absent, mild, moderate and marked[10]. Nasal flaring was noted as either absent or present assessed by observing movement of the nostrils for 30 sec. Pulse rate was counted by palpation and heart-rate per min by the pulse oximeter (Nonin 8500). For blood pressure measurement and pulsus paradoxus mercury sphygnomano-meters with small cuffs (10 cm) were used for the younger children and normal cuffs (15 cm) for the older children. Pulsus paradoxus was defined as a decrease in blood pressure of more than 10 mmHg during inspiration. The SpO2 was measured while the child was breathing room air, with a pulse oximeter (Nonin 8500) using a finger clip on the second finger of the right hand. All the observation about clinical signs were made and recorded by MSR and RPG who were blind to SpO 2 measurement, which was measured by a research staff. In 73% of the cases there was no significant inter-observer difference in various clinical signs measured; in the remaining cases the measurement was repeated and the mean of the two values measured was recorded.
For analysis most symptoms were categorized into a dichotomus scale with values of 0 and 1 table1. Auscultational wheeze was recorded objectively using a scoring system in which children could score between 0-8 points.[10] Wheeze during expiration was categorized as '0' for no wheeze, 1 for end expiratory wheeze, 2 for wheeze during half of the expiration, 3 for wheeze during three-fourth of the expiration and 4 for wheeze during the entire expiration. Likewise, wheezing during inspiration was categorized into 3 categories '0' for no wheeze during inspiration, 1 for wheeze in a part of inspiration and 2 for wheeze during the entire inspiration. Finally, the location of the wheezing sound in the lung was categorized in '0' for no wheeze, 1 for wheeze in 1 or 2 segments of the lung and 2 for wheeze audible in 3 or 4 segments of the lung. By adding up the individual scores of each item, a total wheeze score was acquired. For accessory muscle a child could score 0-9 points[10]. Supra-clavicular, intercostals and sub-costal retractions were categorized into, '0' for none, 1 for mild, 2 for moderate and 3 for marked retractions. By adding up the individual scores of supra-clavicular, intercostals and sub-costal retractions, a total accessory muscle use score was acquired.
Analysis : All the gathered data were analyzed at two time points: at presentation and after one hour. Hypoxemia was defined as a SpO 2 ￡ 92%. Descriptive statistics (mean ± SD) and percentage was used for presentation of data; t-test was used for 'between the groups' comparison of continuous variables and x 2sub -test for discreet variables. The associations between various clinical signs and hypoxemia was evaluated by calculating odds ratios (OR) and 95 percent confidence intervals (95% CI). For this purpose tachypnoea was defined as a respiratory rate > 40/min in children 1 year to 5 years, and > 30/ min in older children. Ordinal scores on auscultational wheezing and accessory muscle use were changed to discrete variable using a cut off, which reflected mild and severe grade of the sign. Cut off score for wheezing was <4 and 3 4, and for accessory muscle use was <3 and 3 3. The variable that were significant on univariate analysis were entered in a multiple logistic regression analysis to identify independent predictors of hypoxemia. All analysis were done using SPSS version 11 and Epi-info 2000.
Results
A total of 51 consecutive patients (34 boys and 17 girls) were assessed at two time points, at admission and after completion of three dosage of salbutamol and budesonide nebulization. The patients were between 29 and 186 months old. (median 82 months and mean ± SD 90.5 ± 38.8 months). The clinical profile of patients at presentation and after one hour is given in table2.
Hypoxemia was seen in 23 (45%) patients at admission and 14 (28.6%) patients after 3 dosage of nebulization. Single breath count, accessory muscle use, dyspnoea, auscultatory wheezing score 34 and pulsus pradoxus were the clinical signs that showed significant association with hypoxemia at presentation and after one hour table3. All the above variables except single breath counting were entered in a logistic regression analysis (backward stepwise) for both the time points using hypoxemia as dependent variable. At both the time points accessory muscle use (score 3 3) and pulsus paradoxus were identified as the two independent predictors of hypoxemia table4. None of the model was highly predictive of hypoxemia table4. Alertness, respiratory rate, auscultational wheezing, cough, pulse, systolic and diastolic blood pressure, and audible wheezing, did not correlate with hypoxemia.
Discussion
Our study shows that at presentation and after initial bronchodilator therapy pulsus paradoxus and accessory muscle use were significantly related to the hypoxemia. Other clinical signs namely single breath count, auscultatory wheeze score 3 4 and fast pulse rate also did reach significance individually but on multiple logistic regression these factors did not stand out as independent predictors of hypoxemia in this population. Single breath count (a reflection of dyspnea) may have reached significance in a larger population sample but it was available in only 50% of the study populations; at younger ages it was not possible to assess this sign properly. Our finding is important since both signs are easy to measure in primary care and may be used as clinical guideposts in the initial assessment of acute childhood asthma. Although both variables have been mentioned in literature as marker of severity of acute asthma, this is the first time that an evidence base is provided in comparison with other clinical variables for evaluation of hypoxemia in acute asthma.
Previous studies on the predictive value of pulsus paradoxus in asthma showed contradictory results.[12],[13],[14],[15] In our study pulsus paradoxus showed a highly significant association with oxygen saturation both at presentation and after 1 hour. It may be considered as a very valuable clinical sign, both in assessing the severity of an asthma attack and predicting hypoxemia.
Impaired alertness and audible wheeze were not common and did not correlate with oxygen saturation at any time. With the exception of one case, none of the patients showed impaired alertness, even when they were hypoxemic. Therefore, alertness and audible wheeze cannot be considered as useful signs in assessing the severity of acute asthma in children. In the way that we categorized the continuous values of respiratory rate, pulse, systolic and diastolic blood pressure and auscultational wheeze none of these symptoms showed any significant correlations with oxygen saturation.
There are some shortcomings in our study. First, we expected to include more patients in our study. Patients came in at a rate of 2-4 per day in April and the inflow decreased to 0-1 per day in June. Nevertheless we believe that the number of patients studied, was sufficient for measuring the association and coming to conclusions. Another possible consequence of the seasonal manner of patient inflow is that there might be a different etiology of asthma in these patients. However, this is considered less relevant to our study, since we were studying the predictability of oxygen saturation by the clinical symptoms in children with asthma, irrespective of the etiology. Our study was conducted in a hospital, which serves both as a referral centre as well as a first contact health care facility. This might result in some selection bias: patients with more severe attacks might have been referred to us. However, since the aim of our study was to assess the symptoms in an acute attack, we did not consider this as a serious limitation to our conclusion and outcome.
We conclude that in assessing an asthma exacerbation, a full clinical assessment of the patient including oximetry is warranted. However, in the absence of a pulse oximeter or peak flow, pulsus paradoxus and accessory muscle use (score 3 3) seems to deserve priority in assessing presence of hypoxemia. We recommend further studies to assess our results in a rural, primary care setting, using a larger sample.
Acknowledgments
We are grateful to the staff of emergency department of PGIMER for their hospitality during our stay in Chandigarh and their help and support during the execution of our study. Special thanks to Dr. Herpreet Kaur Arora and all the patients who participated
References
1. The International Study of Asthma and Allergies in Childhood (ISAAC); Worldwide Variation in Prevalence of symptoms of asthma, allergic rhinoconjuctivitis, and atopic eczema. Lancet 1998; Apr: 1225-1232.
2. Canny GJ, Reisman J, Healy R et al. Acute Asthma: Observations regarding the management in a Pediatric Emergency Room. Paediatrics 1989; 83: 507-512.
3. Li JT, O'Connell EJ. Clinical evaluation of asthma. Pediatr Child Health 1995; 31 : 131-133.
4. Hetzel M, Modell M. Asthma. Occas Pap R Coll Gen Pract 1992; 58 : 14-20.
5. Wright RO, Santucci KA, Jay GD et al. Evaluation of pre- and post-treatment pulse oximetry in acute childhood asthma. Acad Emerg Med 1997; 4 : 114-117.
6. Sole D, Komatsu MK, Carvalho KV, Naspitz CK. Pulse oximetry in the evaluation of the severity of acute asthma and/or wheezing in children. J Asthma 1999; 36 : 327-333.
7. Mehta SV, Parkin PC, Stephens D, Schuh S. Oxygen saturation as a predictor of prolonged, frequent bronchodilator therapy in children with acute asthma. J Pediatr 2004; 145 : 641-645.
8. Keogh KA, Macarthur C, Parkin PC et al. Predictors of hospitalization in children with acute asthma. J Pediatr 2001; 139 : 273-277.
9. Keahey L, Bulloch B, Becker AB et al. Multicenter Asthma Research Collaboration (MARC) Investigators. Initial oxygen saturation as a predictor of admission in children presenting to the emergency department with acute asthma. Ann Emerg Med 2002; 40 : 300-307.
10. Lowell DI, Lister G, Von Kross H et al. Wheezing in infants: the response to epinephrine. Pediatrics 1987; 79 : 939-945.
11. Sethi GR, Choudhary P, Sachdev HPS. Chapter 7- Acute severe asthma. In: Sachdev HPS, Puri RK, Bagga A edrs. Principles of Pediatric and Neonatal Emergencies . Jaypee Brothers. New Delhi, 1994, pp.71-84.
12. Levi M, Hart W, Wieling W. Physical diagnosis - paradoxical pulse. Ned Tijdschr Geneesk 1999; 143 : 2045-2048.
13. Frey B, Freezer N. Diagnostic value and pathophysiologic basis of pulsus paradoxus in infants and children with respiratory disease. Pediatr Pulmonol 2001; 31 : 138-143.
14. Wright RO, Steele DW, Santucci KA et al. Continuous, non-invasive measurement of pulsus paradoxus in patients with acute Asthma. Allergy Asthma Immunol 1996; 76 : 1-13.
15. Chavannes N. Pulse oximetry and respiratory disease in primary care. Prim Care Resp J 2003; 12 : 2-3.(Rahnama'i MS, Geilen RP, )