Impaired gas exchange: accuracy of defining characteristics in childrenwith acute respiratory infection1

Introduction

Nursing diagnoses related to respiratory function, specifically Impaired gas exchange,Ineffective airway clearance, and Ineffective breathing pattern have been frequentlyindicated in the literature as affecting people in different age ranges andsituations1^-6. Of these, Impaired gas exchange is a severeclinical condition defined as an "excess or deficit in oxygenation and/or carbon dioxideelimination at the alveolar-capillary membrane"7.

According to the North American Nursing Diagnosis Association (NANDA-I), this diagnosisbelongs to the domain Elimination and Exchange, Respiratory Function class, and thedefining characteristics of it include: nasal flaring; headache upon awakening; cyanosis(in neonates only); confusion; abnormal skin color (e.g., pale, dusky); diaphoresis;decreased carbon dioxide; dyspnea; visual disturbances; abnormal arterial blood gases;hypercapnia; hypoxia; hypoxemia; restlessness; irritability; abnormal arterial pH;abnormal breathing (e.g., rate, rhythm, depth); somnolence; and tachycardia7.

In acute respiratory infections, such as pneumonia, the functions of gas exchange in thelungs change according to the stages of the disease, resulting in two pulmonary changes:a reduced ratio between ventilation and perfusion and a decrease of the respiratorymembrane's total surface area available. Both situations lead to hypoxemia andhypercapnia, which are defining characteristics of the Impaired gas exchangediagnosis8. Nonethless, even though conditionssuch as acute respiratory infection may lead to this nursing diagnosis, there are fewstudies addressing accuracy concerning this subject.

Acute respiratory infections are most common during childhood and contribute to highlevels of morbidity and mortality among children under the age of five. This is the mostaffected age range due to the susceptibility and immaturity of the respiratory tract atthis age. Respiratory infections are classified as upper or lower respiratory tract,depending on its degree of involvement. Lower respiratory tract infections affect thelower airways and tend to last longer and, if not properly treated, may endanger thechild's life9.

In the face of such a situation, nurses should carefully assess respiratory function toestablish a nursing diagnosis regarding the patient's clinical condition early on andimplement nursing interventions intended to reach its resolution.

Establishing a nursing diagnosis, however, is a process full of uncertainties. For thisreason, nurses should use a diagnostic rationale to find standard signs and symptomscompatible with the most likely diagnoses10. Theidentification of each new defining characteristic may confirm a diagnostic suspicion,eliminate one or redirect the nurse's attention to a human response, not considered tothat point. Thus, studies of diagnostic tests can be used to determine the probabilityof the presence of a given nursing diagnosis11.

Studies contributing to the establishment of useful defining characteristics help tominimize the variability existing in clinical situations presented by the patient and toidentify the nursing diagnosis accurately that represents the true clinical condition.Usually, a single piece of clinical information is not sufficient to confirm thepresence of a nursing diagnosis safely. Hence, a set of defining characteristics needsto be established and the relationship of these with plausible diagnostic hypotheses fora specific clinical situation needs to be verified12.

It is important to note that the prevalence and accuracy measures of the definingcharacteristics of a given nursing diagnosis vary according to the particularities ofthe population under study2^-3^,13^-15. Additionally, the fact that the Impaired gasexchange diagnosis shares defining characteristics that are common to other nursingdiagnoses may make its identification difficult.

Given the previous discussion and aiming to improve nursing diagnoses' rationales, thisstudy's objective was to analyze the accuracy of defining characteristics of theImpaired gas exchange diagnosis among children with acute respiratory infections.

Method

Open prospective cohort study performed with a group of 136 children with acuterespiratory infection for a consecutive period of at least six days and ten days at mostto verify the occurrence of the Impaired gas exchange diagnosis. Prospective cohortsenable the complete and accurate measurement of information concerning clinical signsand symptoms considering temporal dependency among the variables. Due to the shortduration of hospitalization of children with respiratory infections, we opted for anopen cohort in which each individual was included at the time of admission.

The study was conducted in two public children's hospitals located in the Northeast ofBrazil. The study project was approved by the Institutional Review Board of one of thefacilities. The parents or legal guardians were informed of the study's objectives andconsented to data collection by signing free and informed consent forms.

Inclusion criteria were being admitted to the hospital for less than 48 hours and agedfrom zero to five years old. Acute respiratory infections included: pneumonia,bronchiolitis, sinusitis, pharyngitis and tonsillitis diagnosed by the facility'sphysician. Children who did not complete a minimum of six days of follow-up(discontinuity criterion) or had chronic diseases that changed the specific clinicalcondition of acute respiratory infection (e.g., congenital heart disease or cerebralpalsy) were excluded from the study.

The patients were recruited through consecutive sampling as they were admitted to thehospital and after verifying inclusion and exclusion criteria. The sample size wascomputed considering a confidence level of 95%, with a minimum sensitivity of 80%, withconfidence intervals of 13%, and an estimated prevalence of 27.2%, according to a priorstudy1. Based on these values, an estimate of134 children was found. The final sample was composed of 136 children who were assessedfor a consecutive period of six to ten days, so that total number of assessments totaled1,128.

Results

The children assessed in this study were hospitalized for an average of 8.29 days (SD: ±1.58); were 20.35 months old on average (SD: ± 3.11); and 58.1% were boys. The medicaldiagnosis most frequently observed was pneumonia (85.3%), though some children (11.8%)were admitted without having their respiratory infection specified and, in some cases,presented more than one medical diagnosis.

Figure 1 presents the distribution of the nursingdiagnosis Impaired gas exchange and its defining characteristics during the timechildren with acute respiratory infection were monitored. The highest percentage (42.6%)of children with Impaired gas exchange was observed on the day one. On the days two andthree, approximately 38% of children presented this diagnosis, while the number ofchildren with this diagnosis kept decreasing up to the last day of monitoring. In regardto the defining characteristics, abnormal breathing was the most prevalent over thecourse of the ten days of assessment, the values of which ranged from 79.5% to 69.1% onthe first and tenth days, respectively. Apnea was the second most frequentcharacteristic, presenting decreasing percentages over the course of the follow-up:74.3% on the first day and 30.9% on the tenth day. Hypoxemia was frequently observedonly in the first three assessments.

Figure 1.

Distribution of defining characteristics of the diagnosis Impaired gasexchange during the monitoring of children with acute respiratory infection(n=136)

The model of the generalized estimating equation showed that the presence of thecharacteristics restlessness (OR = 11.37), cyanosis (OR = 87.83), abnormal skin color(OR = 10.06), hypoxemia (OR = 11642.1), and abnormal breathing (OR = 6.26) areassociated with a greater likelihood of children with acute respiratory infectionmanifesting the Impaired gas exchange diagnosis when compared to children who did notpresent these characteristics (Table 1).

The accuracy of the defining characteristics that presented statistical significance (p< 0.05) was assessed based on the results obtained through the model of thegeneralized estimating equation. The characteristic that presented the best measures ofaccuracy was hypoxemia (Sensitivity: 96.57%; Specificity: 98.38%; Predictive value+:95.97%; Predictive value-: 93.21%). Other characteristics that also presented a highlevel of accuracy, above 70%, were: abnormal breathing (Sensitivity and Predictivevalue-); restlessness (Specificity and Predictive value-); cyanosis (Specificity,Predictive value+ and Predictive value-); and abnormal skin color (Specificity,Predictive value-). These results are presented in Table 2.

Discussion

Usually, respiratory nursing diagnoses are a priority because they directly affecttissue oxygenation, which is a vital function. Identification of these diagnoses isparticularly important among individuals with diseases in the respiratory tract due toimpairment caused in their airways20. Among thesediagnoses, Impaired gas exchange stands out because it is related to acute respiratoryinfection, which causes changes that negatively impact the functionality of therespiratory system, contributing to the onset of characteristic signs and symptoms.

The prevalence of the diagnosis Impaired gas exchange (42.6%) verified in the firstassessment diverged from another study1 that wasalso conducted with children with acute respiratory infections (27.2%). These divergentresults may be related to the fact that that study was developed in a secondary publichospital, the children of which usually present more stable clinical conditions. Thisstudy, though, was conducted in a tertiary hospital, to which children with more severeconditions are referred, including those coming from the aforementioned secondaryhospital.

No studies addressing Impaired gas exchange using methodology similar to this study werefound. There is, however, one meta-analysis21thataddressed this diagnosis using data concerning the prevalence of definingcharacteristics presented in the literature to determine their accuracy measures.

In this study, the characteristic hypoxemia presented the best measures of accuracy forthe diagnosis Impaired gas exchange, which corroborates data from another studyconducted with children with acute respiratory infections (Sensitivity: 90% andSpecificity: 95%)21. This clinical indicator isassociated with respiratory infections in children due to compromised respiratoryfunction and alveolar ventilation. This occurs due to the retention of secretions, whichmay cause atelectasis, accruing from the occlusion of airways, compromising gas exchangeand triggering hypoxemia22.

Depending on the degree of hypoxemia, imbalance between the supply and demand of oxygenmay lead to the appearance of the defining characteristic cyanosis, which, in thisstudy, presented specificity and positive and negative predictive values. Nonetheless,no studies with statistically significant results were found for the purpose ofcomparison.

In this study, abnormal respiration was a characteristic that presented high sensitivity(82.92%) and negative predictive value (84.42%) to determine Impaired gas exchange inchildren with acute respiratory infections. It is important, however, to highlight thatthe presence of this diagnosis was indirectly determined by the manifestation of atleast one of the defining characteristics, namely: change in rhythm, respiratory rate ordepth.

The literature shows that increased respiratory frequency and depth can occur as thebody's compensatory mechanism attempts to increase the flow of air in the respiratorysystem to fight high levels of carbon dioxide and hydrogen ions in the blood. Thesechanges can be triggered by compromised airways due to the presence of secretionsretained in the airways, which is common in children with respiratory infections23.

Note that even though the defining characteristic abnormal breathing belongs to theImpaired gas exchange diagnosis, it is indirectly related to the definingcharacteristics changes in respiratory depth, tachypnea and bradypnea, which togetherconstitute the Ineffective breathing pattern diagnosis7. Hence, it is possible that nurses have difficulty to precisely inferringrespiratory nursing diagnoses when facing specific clinical situations. It is due to thefact that the process of inferring related nursing diagnoses may be affected when casespresent similar defining characteristics or characteristics with descriptions that leadto the incorporation of information from another characteristic.

This may explain the divergent results reported by a study conducted with adult patientswith pulmonary diseases receiving mechanical ventilation and the results of anotherstudy addressing children with respiratory infections, since it did not presentsignificant statistical values21.

Restlessness presented high specificity in the determination of Impaired gas exchange inthe population under study, corroborating the results of another study conducted with asimilar population, the specificity of which was 91%21. Restlessness may be triggered by conditions that change the respiratorystate, such as acute respiratory infection, and is an important sign in cases in whichthere is hypoxic and respiratory failure. Inappropriate gas exchange intensifies signsof respiratory failure, making breathing a conscious effort, which results inapprehension, agitation and restlessness8.

Nonetheless, no statistically significant relationship was found for the restlessnesscharacteristic in a study conducted with adult patients using invasive ventilationsupport21. It is known that patients usingmechanical ventilation may be kept under sedation, which may compromise theidentification of clinical manifestations of restlessness. Hence, we conjecture thatthis fact might explain the lack of statistical significance.

Abnormal skin color (paleness) stood out due to specificity values and negativepredictive value. Paleness may be related to the generalized mechanism ofvasoconstriction as a consequence of neurogenic or hormonal stimuli and also due todecreased cardiac output, severe anemia, hypovolemia, acidosis or hyportermia24.

The relationship of the abnormal skin color characteristic with the Impaired gasexchange diagnosis may be explained by pathological processes such as pneumonia, whichmay obstruct the airways, trap gases, cause atelectasis, and increase dead space.Consequently, increase in the partial pressure of carbon dioxide in the blood leads torespiratory acidosis and stimulates chemo-sensitive bulb regions (centralchemoreceptors), producing vasoconstriction and increasing peripheral resistance8.

In regard to the prevalence of the defining characteristics under study, dyspnea andabnormal breathing were the most frequently found. A possible explanation is thatdiseases affecting the exchange of respiratory gas through the pulmonaryalveolar-capillary membrane, such as an acute respiratory infection, may promotedisturbances in ventilation/perfusion, with an excess of carbon dioxide and an oxygendeficit. In this way, the body increases breathing as a compensating mechanismattempting to reach normal levels of these gases in the blood8.

The vigorous activity of the respiratory muscles contribute to the onset of abnormalbreathing patterns that may manifest through dyspnea, changes in breathing frequency,rate or depth. These clinical indicators corroborate the results found in a similarstudy, in which these clinical manifestations were reported with high prevalence1.

Additionally, the result obtained by the Model of the generalized estimating equationfor the Impaired gas exchange diagnosis showed that the presence of the characteristicsof restlessness, cyanosis, abnormal skin color, hypoxemia, and abnormal breathing arerelated to an increased likelihood of this diagnosis occurring in children with acuterespiratory infections. Respiratory and alveolar ventilation impairment caused byrespiratory infections trigger adaptive compensatory mechanisms, which if not sufficientto stabilize breathing, may lead to other more severe manifestations.

As previously discussed, hypoxemia caused by an imbalance between oxygen supply anddemand may lead to abnormal breathing, restlessness, cyanosis, or paleness. Therefore,the set of these clinical signs may increase the likelihood that children with acuterespiratory infection develop Impaired gas exchange. The close relationship of hypoxemiawith this diagnosis was observed by the accuracy measures obtained.

It is important to note that the scarcity of studies in the literature with amethodological design similar to this study limited the comparison of results.Therefore, we believe that similar studies addressing children with acute respiratoryinfections are needed to enable comparison with these findings. Note that the resultspresented here may have been influenced by the bias of incorporation and diagnosticassessment, which happens when prior knowledge concerning defining characteristics isincorporated during the process of diagnostic inference25.

Despite the contribution of this study's findings concerning the accurate identificationof the Impaired gas exchange diagnosis among children with acute respiratory infections,the results should be addressed with care, as most children were assessed in a hospitalthat provides care to patients with a greater probability of manifesting more severeclinical conditions.

Conclusion

The Impaired gas exchange nursing diagnosis was manifested in 42.5% of the sample. Themost prevalent defining characteristics were abnormal breathing, dyspnea and hypoxemia.The model of generalized estimating equations showed that the concomitant presence ofrestlessness, cyanosis, abnormal skin color, hypoxemia and abnormal breathing areassociated with an increased likelihood of the occurrence of this diagnosis in childrenwith acute respiratory infections.

In regard to accuracy measures, hypoxemia was the defining characteristic that predictedthe occurrence of Impaired gas exchange diagnosis. Other characteristics, however, alsopresented high levels of sensitivity (abnormal breathing) and specificity (restlessness,cyanosis, and abnormal skin color).

It is believed that the determination of the predictive ability of these definingcharacteristics increases the reliability of the process of diagnostic inference andenables nurses to produce hypotheses consisting of more likely nursing diagnoses in therepresentation of the clinical situation presented by the patient.