Резюме
Махаринська О.С., Іманова Н.І. Можливості прогнозування результатів лікування негоспітальної пневмонії у хворих на цукровий діабет 2 типу за допомогою запитальників  CapSym-12 та «Шкала оцінки НП» El Moussaoui R.
У сучасній світовій літературі існує багато моделей прогнозування клінічного кінця негоспітальної пневмонії (НП), однак жодна з них не включає дані запитальників для хворих на цукровий діабет 2 типу (ЦД). Метою нашого дослідження було визначити чи існує можливість прогнозування клінічного кінця захворювання за допомогою результатів запитальників, що визначають якість життя та динаміку симптомів НП у хворих на ЦД 2 типу. Для досягнення поставленої мети було обстежено 82 пацієнта у віці від 51 до 82 років, хворих на НП та супутній ЦД 2. Динаміку показників якості життя пацієнтів та динаміку симптомів НП оцінювали за допомогою обраних нами запитальників CapSym-12 та «Шкала оцінки НП» El Moussaoui R (ШОПН). Пацієнтів опитували в день госпіталізації (день «0»), через 48–72 години і 10±2 днів після початку лікування. Пацієнти відповідали на запитальник самостійно або за допомогою лікаря, що призначав лікування. За допомогою логістичної регресії нами було встановлено вибірку статистично важливих показників динаміки симптомів запитальників, прогностичну можливість котрих використали у нашому дослідженні задля оцінки стану здоров’я і динаміки симптомів НП у хворих на ЦД 2 і які дозволяють спрогнозувати результат лікування НП. Результати лікування НП у хворих ми оцінювали як: одужання, поліпшення, невдача, неможливо оцінити. Одним із математичних методів оцінки діагностичної цінності і співвідношення між чуттєвістю і специфічністю тесту є побудування характеристичних ROC-кривих і обчислення площі під характеристичною кривою. Розраховані у нашому дослідженні ROC-криві дозволили спрогнозувати результат лікування «невдача» у хворих на ЦД 2, що були госпіталізовані із діагнозом НП. Таким чином, комплексний аналіз клінічної картини захворювання, доповнений даними анамнезу захворювання і результатами сучасних запитальників на НП, є раціональним підходом до рішення питань оцінки результату лікування НП у хворих на ЦД 2.
Ключові слова: негоспітальна пневмонія, цукровий діабет 2 типу, запитальники, прогноз, ROC-крива.
Резюме
Махаринская Е.С., Иманова Н.И. Возможности прогнозирования результатов лечения внебольничной пневмонии у больных сахарным диабетом 2 типа с помощью вопросников CapSym-12 и «Шкала оценки ЧП» El Moussaoui R.
В современной мировой литературе существует много моделей прогнозирования клинического исхода внебольничной пневмонии (ВП), однако ни одна из них не включает данные результатов вопросников для больных с сахарным диабетом 2 типа (СД). Целью нашего исследования было определить существует возможность прогнозирования клинического исхода заболевания с помощью результатов вопросников, определяющих качество жизни и динамику симптомов ВП у больных с СД 2 типа. Для достижения поставленной цели было обследовано 82 пациента в возрасте от 51 до 82 лет, больных ВП и сопутствующим СД 2 типа. Динамику показателей качества жизни пациентов и динамику симптомов ВП оценивали с помощью выбранных нами вопросников CapSym-12 и «Шкала оценки ВП» El Moussaoui R (ШОВП). Пациентов опрашивали в день госпитализации (день «0»), через 48-72 часа и 10 ± 2 дня после начала лечения. Пациенты отвечали на вопросник самостоятельно или с помощью врача, который назначал лечение. С помощью логистической регрессии нами было установлена выборка статистически важных показателей динамики симптомов опросников, прогностическую возможность которых использовали в нашем исследовании для оценки состояния здоровья и динамики симптомов ВП у больных с СД 2, которые позволяли прогнозировать результат лечения ВП. Результаты лечения ВП у больных мы оценивали как: выздоровление, улучшение, неудача, невозможно оценить. Одним из математических методов оценки диагностической ценности и соотношения между чувствительностью и специфичностью теста является построение характеристических ROC-кривых и вычисление площади под характеристической кривой. Рассчитанные в нашем исследовании ROC-кривые позволили спрогнозировать результат лечения «неудача» у больных с СД 2, которые были госпитализированы с диагнозом ВП. Таким образом, комплексный анализ клинической картины заболевания, дополненный данными анамнеза заболевания и результатами современных вопросников по изучению качества жизни и динамики симптомов ВП, является рациональным подходом к решению вопросов оценки результата лечения ВП у больных с СД 2.
Ключевые слова: внебольничная пневмония, сахарный диабет 2 типа, вопросники, прогноз, ROC-кривая.
Summary
Makharynska О.S., Imanova N.I. The possibility of predicting outcomes community-acquired pneumonia in patients with type 2 diabetes using questionnaires CapSym-12 and CAP El Moussaoui R.
In the modern world literature there are many models, using that can to predict clinical end of CAP, but none of them includes the data results of the questionnaires for patients with type 2 diabetes. The aim of our study was to determine there is a possibility of predicting clinical outcome of the disease with the help of questionnaires, the results of which determine the quality of life and the dynamics of the symptoms of CAP in patients with type 2 diabetes. To achieve this goal were examined 82 patients with type 2 diabetes aged 51 to 82 years, which were hospitalized with CAP. The dynamics of indicators of quality of life and the dynamics of symptoms was assessed using the CAP we selected questionnaires CapSym-12 and CAP El Moussaoui R. Patients were interviewed at the day of admission (day «0»), after 48-72 hours and 10±2 days after treatment. Patients answered a questionnaire on their own or with the help of a doctor who administered the treatment. Using logistic regression, we established a statistically significant sample of the dynamics of symptoms questionnaire, predictive capability that used in our study to assess the health and dynamics of the symptoms of CAP in patients with type 2 diabetes and are predictive of treatment outcome for patients with CAP. The results of the treatment of CAP patients we evaluated as cure, improvement, failure and can not be estimated. We used the mathematical methods to assess the diagnostic value and the ratio between the sensitivity and specificity of the test is to construct a characteristic ROC-curves and calculating the area under the ROC curve. Calculated in our study, ROC-curves are allowed to predict the outcome of the treatment of ‘failure’ in patients with type 2 diabetes who were hospitalized with a diagnosis of CAP. Thus, a comprehensive analysis of the clinical picture of the disease, complete case history of the disease and the results of modern questionnaires to study the quality of life and symptoms and signs to the CAP, is a rational approach to issues of evaluation of the treatment of CAP in patients with type 2 diabetes.
Key words: community-acquired pneumonia, diabetes type 2, questionnaires, forecast, ROC-curve.

УДК 616.24-002-085.23-06:616.379-008.64

Харьковская медицинская академия последипломного образования

61176, г. Харьков, ул.Корчагинцев, 58

Kharkov medical academy of Postgraduate Education

61176, Kharkov, 58 str.Korchahyntsev

grimoire@rambler.ru

One of the leading roles in increasing the incidence and mortality of CAP (community acquired pneumonia), despite some progress in its treatment continues to play DM (diabetes mellitus type 2). The presence of this disease increases the risk of adverse effects to patients with CAP [1, 4]. In the modern world literature there are many models, which can be useful to predict clinical end state of emergency, but none of them includes data questionnaires that assess quality of life and symptom dynamics of CAP.
The aim of our study was to determine whether it is possible to predict the end of the CAP disease by questionnaire’s results, which determine the quality of life and the dynamics of CAP symptoms in patients with type 2 diabetes.
Materials and methods. To achieve this goal were examined 82 patients aged 51 to 82 years, with CAP and concomitant DM type 2. The dynamics of quality of life for patients and dynamics of CAP symptoms was assessed using questionnaires selected by us: CapSym-12 and CAP El Moussaoui R (SHOPN). Patients were interviewed on the day of admission (day "0"), 48-72 hours, and 10 ± 2 days after the start of treatment. Patients answered a questionnaire on their own or with the help of a doctor, who administrated the treatment. Questionnaire CapSym-12 [5] was developed in English using the "gold standard" and translated and validated on 12 other languages ​​according to the appropriate methodology. With the consent of the patients in the study we used the Russian version of the questionnaire. It allows to measure twelve related emergency symptoms: cough, chest pain, shortness of breath, sweating, cool, headache, nausea, muscle pain, lack of appetite, ability to concentrate, insomnia, fatigue. Patients were asked to self or by a physician to evaluate bothersome of each symptom in the last 24 hours with 6-level Likert-scale (0 = the patient had no symptoms, 1 = symptoms bothered patients, 2 = somewhat worried, 3 = moderate, 4 = really worried, 5 = very worried). Questionnaire CAP El Moussaoui R. (SHOPN) [6] was developed in English and translated into Russian and Ukrainian languages; it was validated by the department of internal medecine of KMAPO [3]. It allows us to estimate 8-related emergency symptoms (presence and severity of dyspnea, presence and severity of cough, presence and character of sputum expectoration if available) and 2 symptoms of the quality of life of patients (patient feels healthy? and overall health ), ranked by Likert-scale. With the consent of the patients in the study used the Russian version of the questionnaire. Replies of patients to questionnaire items were calculated using the rating scale CAP (topics: CAP-rate, respiratory scale and the rate of wellbeeing).
Results and discussion: Using logistic regression, we found a statistically significant sample of the dynamics of symptom questionnaires, predictive ability which used in this study to assess the health and dynamics of CAP symptoms in patients with type 2, which allows to predict the outcome of emergency. Predictive value of the test - the probability presence (absence) of disease in the known results of the study. Predictive value of a positive result - the probability of the presence of disease in the positive (abnormal) result of the diagnostic test. Predictive value of a negative result - the probability of disease in the presence of negative (normal) result of the diagnostic test. Predictability depends on the sensitivity and specificity of the diagnostic method of spread of the disease in the population that was studied. More sensitive test gives the higher predictive value of the negative result. What is a more specific test, the higher the predictive value of a positive result [2].
     Results of treatment of emergency patients we evaluated as cure, improvement, failure, can not be estimated. One of mathematical methods to assess the diagnostic value and the relationship between sensibility and specificity of the test is building a characteristic ROC-curves and calculating the area under the characteristic curve. ROC-curve shows the relationship between sensibility and specificity of the method of analysis, the area under the ROC-curve defines an objective indicator of the quality of the classifier, which allows you to compare the curves together.
     During the assessment of performance questionnaires on the day of hospitalization, 48-72 hours and 10 ± 2 of treatment significant results were received. However, assessing the impact of their dynamics of questionnaires during the treatment period we have identified significant positive relationship dynamics prognostic index and respiratory emergency scale CAP-questionnaire between day "48-72 hours" of treatment and day of hospitalization (day "0") on the outcome CAP "recovery." Also predicted "recovery" emergency patients with concomitant type 2 allows the dynamics of emergency index CAP-questionnaire between days of observation "after 10 ± 2 of treatment" and the day of admission. The negative results of the questionnaire CAP-Sym-12 in the table due to the fact that the dynamics in the improvement of patient questionnaire total score decreases, whereas in the case of the CAP-questionnaire improvement of the patient at the emergency rate increases and questionnaire. Negative treatment outcome was considered as a "failure" and "improvement." None of the indicators of questionnaires was not significantly associated with the variable "outcome improvement." Variable “outcome failure” according to the analysis was associated with indicators questionnaire CAP-Sym-12 through 48-72 hours of starting treatment and the dynamics of the questionnaire CAP-Sym-12 between the day" after 48-72 hours of treatment " and day "0", and the dynamics of the well beeing scale of CAP-questionnaire between day "after 10 ± 2 days of starting treatment" and day "0". Provided that the positive result was considered treatment as a "cure" and "improvement."
 Calculated in our study, ROC-curves for the outcome of treatment "failures" are presented in figures in the table below, where:
- CAP13 - ROC-curve index questionnaire emergency SHONP between days of observation, "48-72 hours of starting treatment" and day "0", Square ROC-scale 0.64,p<0.037;
- Well13 - ROC-curve scale health questionnaire SHONP between days of observation "after 10 ± 2 days of starting treatment" and day "0", Square ROC-scale 0.59,p<0.05;
- Opr2 (3) - ROC-curve questionnaire CAP-Sym-12 day observation: "48-72 hours of starting treatment"; area ROC-scale 0.59,p<0.03;
- Opr2 (13) - ROC-curve results of the questionnaire CAP-Sym-12 between days of observation, "48-72 hours of starting treatment" and day "0", Square ROC-scale 0.71,p<0.05.
   Clinical observation
Patient K., 69 years old, was treated in the therapeutic department of city hospital number 25 in Kharkiv from 24.12.2007 to 09.01.2008 years (duration of treatment was 16 days) with a diagnosis of community acquired pneumonia in right -low lobe, secondary severity. Type 2 diabetes, moderate, compensated. Coronary artery disease. Hypertension 2. HF IIA, II f.c. according to NYHA.
Befor hospitalization within 1 week, the diagnosis was established by district doctor, he received the appointment of b-lactams orally. Treatment hadn’t a positive effect: the temperature rose to 39 ° C, started strong cough. Patinet was admitted the hospital number 25 in Kharkiv.
Suffering from diabetes since 2002, takes regular "gliklazid".
During hospitalization he had complains of fever, presence of cough with sputum, weakness. On the day of admission the patient was moderate. The body temperature of 37,6 ° C. Normal skin color, clean. During auscultation of the lungs listened hard breathing in the lower regions of the right - crackling. RR 20 per minute. Cardiac rhythmic, muffled. Heart rate - 84 per minute. Blood pressure 140/50 mm Hg. century. The abdomen during palpation soft and painless. The liver was not enlarged. Symptom Pasternatskogo was negative. No oedema.
Chest radiograph on 25.12.2007: the lower and middle fractions - infiltration of lung tissue medium intensity. Roots of the lungs expanded. Conclusion: community acquired pneumonia in the lower right lobe.

Red Blood 24.12.2007 was: erythrocytes - 5,2 * 1012 / L, WBC - 5.0 * 109 / L, ESR - 5 mm / h, stab neutrophils - 1% segmented neutrophils - 69%, lymphocytes - 25%, monocytes - 4%, eozinofily - 1%.
Glycemic profile during treatment: 11.0 - 12.0 - 5.5 mmol / l.
Clinical analysis of sputum from 27.12. and 28.12. 2007: character - mucose-purulent, white blood cells - a bit, erythrocytes - 1-2 p / view.
From 26.12 USound Conclusion: Atherosclerosis of the aorta. FV-54%. Dilatation of the left atrium. Left ventricular hypertrophy 1 degree. Fibrosis of the aortic and mitral valves. Hypertrophy of the right ventricle. Fibrosis of the pancreas. Left renal sinus cysts.
On a scale of CRB-65 patients during hospitalization was 2 points (age> 65 years, DBP <60 mmHg).
The patient was prescribed therapy: respiratory fluoroquinolone livofloksatsyn 500 mg / day from 24.12 to 30.12. 2007. Concomitant drug therapy: lizinopril, ketorolac, ambroxol, atorvastatin.
After 48-72 hours the patient remained without positive dynamics. The patient complained of severe general weakness, cough with sputum yellow-green. Patient's body temperature was 38,4 ° C. Auscultation of the lungs: hard breathing in the lower regions, on the right - crackling. BR 22 per minute. Cardiac rhythmic, muffled. Heart rate - 78 per minute. Blood pressure was 130/60 mm Hg. The abdomen was normal. No edema. To treatment was added b - lactam: proksetil cefpodoxime 200 mg 2 times a day orally from 26.12.07. to 06.01.2008, the total duration of antibacterial treatment was 14 days.
Red Blood 28.12.2007 was: erythrocytes - 5,5 * 1012 / L, WBC - 3.0 * 109 / L, ESR - 78 mm / h, stab neutrophils - 9% segmented neutrophils - 68%, lymphocytes - 28%, monocytes - 3% eosinophil - 1%.
After 10 ± 2 days of treatment the patient's condition improved significantly. The patient complained of general weakness, loose cough with yellowish sputum release. Body temperature - 37,4 ° C. Auscultation of the lungs: hard breathing in the lower regions, on the right - crackling. BR 20 per minute. Cardiac rhythmic, muffled. Heart rate - 82 per minute. Blood pressure 140/80 mm Hg. The abdomen was normal. No oedema.
Chest radiograph on 08.01.2008: the segment of infiltration in S9 of lung tissue under reverse development. Conclusion: The residual of community-acquired pneumonia.
After 15 days the patient improved, complaints on liquid cough with phlegm mucus release. Body temperature - 36,4 ° C. Auscultation of the lungs: hard breathing. BR 18 per minute. Cardiac rhythmic, muffled. Heart rate - 68 per minute. Blood pressure was 130/80 mm Hg. The abdomen was normal. No oedema. The patient was discharged from the department under the supervision of the district physician. Results of treatment of patients defined as "failure." Combined treatment of mild state of emergency by b-lactam and respiratory fluoroquinolones in Ukraine is quite common [7].
The results of patient questionnaires:
Dynamics of Questionnaires

After 48-72 hours CAP-Sym-index = 12, 27.0
Day "0" - 48-72 hours emergency Index 6.0
Respiratory Scale 33.4
Scale being 15.3
CAP-Sym-12 -5
Day "0" - after 10 ± 2 days of emergency Index 9.5
Respiratory Scale 11.1
Scale being 5.7
CAP-Sym-12 -11
    Thus, before the end of the treatment period as a result of questionnaires CAP-Sym-12 and CAP and analysis of the dynamics of one could assume negative emergency treatment in patients with concomitant diabetes. High rate of CAP-Sym-12 at the start of treatment (cut-off point = 28,4 points), a small speaker and between the day of admission and after 48-72 hours of treatment (<8.2), low dynamic og scale well beeing questionnaire CAP during the treatment period (<18.9 48-72 hours of starting treatment and <28.1 after 10 ± 2 days).
Conclusions: Thus, a comprehensive analysis of clinical disease, coupled with a history of disease and the results of the questionnaire on current state of emergency, a rational approach to solving the assessment result of emergency treatment in patients with type 2.

Література

1. Внебольничная пневмония у взрослых: практические рекомендации по диагностике, лечению и профилактике: пособие для врачей / Российское респираторное общество. – Москва, 2010. – 106 с.
2. Колосов В.П. Внебольничная пневмония (клиническое течение, прогнозирование исходов) / В.П. Колосов, Е.Ю. Кочегарова, С.В. Нарышкина. - Благовещенск, 2012 –124 с.
3. Махаринська О.С. Застосування та валідизація короткого запитальника для хворих на негоспітальну пневмонію в Україні / О.С. Махаринська, М.М. Лебединська, І.Г. Березняков // Болезни и антибиотики. – 2012. - № 1(6). – С.51-58.
4. Almirall J. New evidence of risk factors for community-acquired pneumonia: a population-based study / J. Almirall, I. Bolıbar [et al.] // Eur Respir J. – 2008. – Vol. 31. – P. 1274–1284.
5. Lamping D.L. The community-acquired pneumonia symptom questionnaire: a new, patient-based outcome measure to evaluate symptoms in patients with community-acquired pneumonia / Donna L. Lamping, Sara Schroter, Patrick Marquis [et al.] // Chest. – 2002. – Vol. 122. – Р. 920-929.
6. Moussaoui Е.R. Long-term symptom recovery and health-related quality of life in patients with mild-to-moderate-severe community-acquired pneumonia / Rachida el Moussaoui, Brent C. Opmeer, Corianne A.J.M. de Borgie [et al.] // Chest. – 2006. - Vol. 130. – Р.1165-1172.
7. Wilson B.Z. Comparison of two guideline-concordant antimicrobial combinations in elderly patients hospitalized with severe community-acquired pneumonia / B.Z. Wilson, A. Anzueto, M.I. Restrepo [et al.] // Crit Care Med. – 2012. - Vol. 4. – P.293-299.