Tesis

Uji Diagnostik Sepsis Awitan Lambat pada Neonatus di Rumah Sakit Cipto Mangunkusumo Jakarta = Diagnostic Test of Late Onset Neonatal Sepsis at Cipto Mangunkusumo Hospital Jakarta.

Latar belakang: Sepsis neonatorum awitan lambat (SNAL) merupakan penyebab utama mortalitas pada neonatus. Kesulitan diagnosis dini SNAL menyebabkan perlunya diagnosis berdasarkan gabungan anamnesis faktor risiko (usia gestasi kurang bulan, berat lahir rendah, perawatan di ruang intensif yang lama, tanpa pemberian kolostrum, lama ventilasi mekanik, lama kateter vena sentral, lama nutrisi parentetal total, dan riwayat operasi saluran cerna), pemeriksaan fisis (instabilitas suhu, gangguan keadaan umum, gangguan sirkulasi, gangguan tonus, dan gangguan pernapasan), dan pemeriksaan laboratorium (jumlah leukosit, absolute neutrofil count, jumlah trombosit, I/T ratio, CRP, dan PCT). Tujuan: Mengetahui hubungan anamnesis faktor risiko, pemeriksaan fisis, dan pemeriksaan laboratorium dengan SNAL, dan mengetahui anamnesis faktor risiko, pemeriksaan fisis, dan pemeriksaan laboratorium yang dapat digunakan sebagai uji diagnostik SNAL. Metode: Uji diagnostik secara potong lintang dengan data sekunder pada pasien neonatus yang dirawat di RSCM selama 1 Januari hingga 31 Desember 2015 dengan diagnosis tersangka SNAL. Hasil penelitian: Sebanyak 113 subyek SNAL memenuhi kriteria inklusi. Variabel tanpa pemberian kolostrum [OR 4,87 (IK95% 1,02; 23,65)], lama kateter vena sentral lebih dari 14 hari [OR 4,34 (IK95% 1,46; 12,88)], trombositopenia [OR 5,94 (IK95% 2,51; 14,04)], peningkatan CRP [OR 6,63 (IK95% 2,46; 17,91)], dan peningkatan PCT [OR 3,34 (IK95% 1,28; 8,76)] terbukti memiliki hubungan bermakna dengan SNAL. Hasil akhir hubungan anamnesis faktor risiko dan laboratorium dengan SNAL dari analis regresi logistik dan metode backward adalah berat lahir rendah, tanpa pemberian kolostrum, lama kateter vena sentral lebih dari 14 hari, trombositopenia, dan peningkatan CRP. Sensitivitas tertinggi yaitu tanpa pemberian kolostrum (96,4%). Spesifisitas tertinggi yaitu berat lahir rendah (77,6%). Positive predictive value tertinggi yaitu lama kateter vena sentral lebih dari 14 hari (76,2%). Negative predictive value tertinggi yaitu tanpa pemberian kolostrum (81,8%). Likelihood ratio + terbaik yaitu trombositopenia (1,9). Likelihood ratio – terbaik yaitu tanpa pemberian kolostrum dan peningkatan CRP (0,2). Nilai area under curve anamnesis faktor risiko (berat lahir rendah, tanpa pemberian kolostrum, lama kateter vena sentral lebih dari 14 hari) digabung dengan trombositopenia dan peningkatan CRP sebesar 82,4% (IK95% 74,9; 89,9%). Kesimpulan: Kombinasi berat lahir rendah, tanpa pemberian kolostrum, lama kateter vena sentral lebih dari 14 hari, trombositopenia, dan peningkatan CRP dapat digunakan sebagai uji diagnostik SNAL.
Kata kunci: Uji diagnostik; sepsis neonatorum awitan lambat.



Background: Late onset neonatal sepsis (LONS) was the primary cause of neonatal mortality in neonates. The difficulties of early diagnosis of LONS lead to diagnosis based on the combination of history of risk factors (prematurity, low birth weight, prolonged admisssion in intensive care, without giving colostrum, duration of mechanical ventilation, duration of central venous catheter, duration of parentetal nutrition, and history of gastrointestinal operation), physical examination (temperature instability, impaired general condition, circulatory and hypotonia, and breathing difficulties), and laboratory tests (leukocyte count, absolute neutrophil count, platelet count, I/T ratio, CRP and PCT). Objective: To determine the relationship of history of risk factors, physical examination, and laboratory tests with LONS, and to know history of risk factors, physical examination, and laboratory tests that could be used as diagnostic test. Methods: Diagnostic test with a cross-sectional study was performed with secondary data on neonatal patients of suspected LONS in RSCM during January 1 st to December 31 st 2015. Results: A total of 113 subjects met the inclusion criteria. Variable without giving colostrum [OR 4.87 (IK95% 1.02; 23.65)], use of central venous catheter more than 14 days [OR 4.34 (IK95% 1.46; 12.88)], thrombocytopenia [OR 5.94 (IK95% 2.51; 14.04)], elevated CRP [OR 6.63 (IK95% 2.46; 17.91)], and elevated PCT [OR 3.34 (IK95% 1.28; 8.76)] had significant relationship with blood culture proven LONS. The final result of relationship of risk factor anamnesis and laboratory with LONS from logistic regression and backward method were low birth weight, without giving colostrum, use of central venous catheters more than 14 days, thrombocytopenia, and elevated CRP. The highest sensitivity was without giving colostrum (96.4%). The highest specificity was low birth weight (77.6%). The highest positive predictive value was long term use of central venous catheter (76.2%). The highest of negative predictive value was without giving colostrum (81.8%). The best of likelihood ratio + was platelets count (1.9). The best of likelihood ratio – was without giving colostrum and elevated CRP (0.2). The value of area under the curve anamnesis of risk factors (low birth weight, without giving colostrum, use of central venous catheters more than 14 days) combined with thrombocytopenia and elevated CRP was 82.4% (IK95% 74.9%; 89.9%). Conclusion: The combination of low birth weight, without giving colostrum, use of central venous catheters more than 14 days, thrombocytopenia, and elevated CRP can be used as a diagnostic test of LONS.
Keyword: Diagnostic test; late onset neonatal sepsis.