Multi-Step Internet Traffic Forecasting Models with Variable Forecast Horizons for Proactive Network Management.

The ISP (Internet Service Provider) industry relies heavily on internet traffic forecasting (ITF) for long-term business strategy planning and proactive network management. Effective ITF frameworks are necessary to manage these networks and prevent network congestion and over-provisioning. This study introduces an ITF model designed for proactive network management. It innovatively combines outlier detection and mitigation techniques with advanced gradient descent and boosting algorithms, including Gradient Boosting Regressor (GBR), Extreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGB), CatBoost Regressor (CBR), and Stochastic Gradient Descent (SGD). In contrast to traditional methods that rely on synthetic datasets, our model addresses the problems caused by real aberrant ISP traffic data. We evaluated our model across varying forecast horizons-six, nine, and twelve steps-demonstrating its adaptability and superior predictive accuracy compared to traditional forecasting models. The integration of the outlier detection and mitigation module significantly enhances the model's performance, ensuring robust and accurate predictions even in the presence of data volatility and anomalies. To guarantee that our suggested model works in real-world situations, our research is based on an extensive experimental setup that uses real internet traffic monitoring from high-speed ISP networks.

DRL-GAN: A Hybrid Approach for Binary and Multiclass Network Intrusion Detection.

Our increasingly connected world continues to face an ever-growing number of network-based attacks. An Intrusion Detection System (IDS) is an essential security technology used for detecting these attacks. Although numerous Machine Learning-based IDSs have been proposed for the detection of malicious network traffic, the majority have difficulty properly detecting and classifying the more uncommon attack types. In this paper, we implement a novel hybrid technique using synthetic data produced by a Generative Adversarial Network (GAN) to use as input for training a Deep Reinforcement Learning (DRL) model. Our GAN model is trained on the NSL-KDD dataset, a publicly available collection of labeled network traffic data specifically designed to support the evaluation and benchmarking of IDSs. Ultimately, our findings demonstrate that training the DRL model on synthetic datasets generated by specific GAN models can result in better performance in correctly classifying minority classes over training on the true imbalanced dataset.

Publications in paediatric critical care medicine: Trends from Pakistan.

Objective: To analyse the characteristics of research published from Pakistan on paediatric critical care medicine. METHODS: The exploratory study was conducted at the Aga Khan University, Karachi from July 2021 to March 2022, and comprised a comprehensive search on MedLine, Google Scholar and PakMediNet databases for literature from Pakistan pertaining to paediatric critical care medicine published between January 2010 and December 2021. The search was done using appropriate key words. Conference abstracts and papers authored by paediatric intensivists with unrelated topics were excluded. Data was extracted on a structured spreadsheet, and was subjected to bibliometric analysis. Data was analysed using SPSS 20. RESULTS: Of the 7,514 studies identified, 146(1.94%) were analysed. These were published in 51 journals with a frequency of 13.3 per year. There were 107(73.3%) original articles, 96(65.8%) were published in PubMed-indexed journals, and 35(24%) were published in locally indexed journals. Further, 100(69.4%) papers were published from 5 paediatric intensive care units in Karachi, and 81(56%) were contributed by a single private-sector hospital. The total citation count was 1072, with 2(1.4%) papers receiving >50 citations. There was a linear trend with some skewing and an annual growth rate of >15%. Conclusion: Publications from Pakistan related to paediatric critical care medicine showed positive linear growth. There was a paucity of multicentre studies, randomised controlled trials, and high-impact publications.

Evaluation of the Resources and Inequities Among Pediatric Critical Care Facilities in Pakistan.

OBJECTIVES: To evaluate nationwide pediatric critical care facilities and resources in Pakistan. DESIGN: Cross-sectional observational study. SETTING: Accredited pediatric training facilities in Pakistan. PATIENTS: None. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A survey was conducted using the Partners in Health 4S (space, staff, stuff, systems) framework, via email or telephone correspondence. We used a scoring system in which each item in our checklist was given a score of 1, if available. Total scores were added up for each component. Additionally, we stratified and analyzed the data between the public and private healthcare sectors. Out of 114 hospitals (accredited for pediatric training), 76 (67%) responded. Fifty-three (70%) of these hospitals had a PICU, with a total of 667 specialized beds and 217 mechanical ventilators. There were 38 (72%) public hospitals and 15 (28%) private hospitals. There were 20 trained intensivists in 16 of 53 PICUs (30%), while 25 of 53 PICUs (47%) had a nurse-patient ratio less than 1:3. Overall, private hospitals were better resourced in many domains of our four Partners in Health framework. The Stuff component scored more than the other three components using analysis of variance testing ( p = 0.003). On cluster analysis, private hospitals ranked higher in Space and Stuff, along with the overall scoring. CONCLUSIONS: There is a general lack of resources, seen disproportionately in the public sector. The scarcity of qualified intensivists and nursing staff poses a challenge to Pakistan's PICU infrastructure.

Towards an Optimized Ensemble Feature Selection for DDoS Detection Using Both Supervised and Unsupervised Method.

With recent advancements in artificial intelligence (AI) and next-generation communication technologies, the demand for Internet-based applications and intelligent digital services is increasing, leading to a significant rise in cyber-attacks such as Distributed Denial-of-Service (DDoS). AI-based DoS detection systems promise adequate identification accuracy with lower false alarms, significantly associated with the data quality used to train the model. Several works have been proposed earlier to select optimum feature subsets for better model generalization and faster learning. However, there is a lack of investigation in the existing literature to identify a common optimum feature set for three main AI methods: machine learning, deep learning, and unsupervised learning. The current works are compromised either with the variation of the feature selection (FS) method or limited to one type of AI model for performance evaluation. Therefore, in this study, we extensively investigated and evaluated the performance of 15 individual FS methods from three major categories: filter-based, wrapper-based, and embedded, and one ensemble feature selection (EnFS) technique. Furthermore, the individual feature subset's quality is evaluated using supervised and unsupervised learning methods for extracting a common best-performing feature subset. According to our experiment, the EnFS method outperforms individual FS and provides a universal best feature set for all kinds of AI models.

Red blood cell transfusion in critically-ill children and its association with outcome.

OBJECTIVE: To determine the indications and threshold of haemoglobin levels for packed red blood cell transfusion and its association with outcomes in a paediatric intensive care setting. METHODS: The retrospective study was conducted in the paediatric intensive care unit of the Aga Khan University Hospital, Karachi, and comprised medical records of all inpatients with age between 1 month and 16 years who received packed red blood cell transfusions between January and December 2017. Data was retrieved from the hospital database and was analyzed using SPSS 22. RESULTS: Of the 147 subjects with a mean age of 67.89±65.8 months, 76(51.7%) were males. Mean paediatric risk of mortality score was 11.72±7.86. Major admitting diagnosis included sepsis and multiorgan dysfunction 50(34%), respiratory diseases 26(17.7%) and haematology/oncology diseases 22(15%). The indications for transfusion was low haemoglobin in 90(61.2%) patients, shock 29(19.7%) and hypoxia 28(19%). Acute transfusion reaction was observed in 1(0.7%) patient; 120(82%) required mechanical ventilation; and 94(64%) required inotropic support. Of the total, 88(59.9%) patients survived. Paediatric risk of mortality score, need for inotropic support and mechanical ventilation were associated with mortality (p<0.05). CONCLUSIONS: Packed red blood cell transfusion, which is frequently prescribed in intensive care settings, was not found to be associated with favourable outcome.

Gastrointestinal Complications in Critically Ill Children: Experience from A Resource-Limited Country.

OBJECTIVES: To determine the frequency and predictors of outcome of gastrointestinal complications (GIC) in critically ill children. METHODS: This descriptive study was prospectively conducted in The Pediatric Intensive Care Unit (PICU), The Aga Khan University Hospital (AKUH), Karachi, from September 2015 to January 2017. After obtaining approval from the Ethical Review Committee of AKUH and informed consent from the parents, all children (aged one month to 18 years), of either gender, admitted to the Pediatric Intensive Care Unit (PICU) during the study period were included. The frequency of the defined GIC: vomiting, high gastric residue volume (GRV), diarrhea, constipation, and gastrointestinal bleed were recorded daily for the first week of the PICU stay. The data was collected by the primary investigator on a predesigned data collection form with inclusion of variables and predictors in light of existing literature and local expertise. The questionnaire was shared with the Pediatric Critical Care Medicine faculty and a consensus was sought on the elements to be incorporated. RESULTS: GIC developed within the first 48 hours of admission in 78 (41%) patients. Of the patients who developed GIC, 37 (47.4%) patients developed high GRV: 31 (39.7%) patients developed constipation, 18 (23.1%) patients developed vomiting, 14 (17.9%) patients developed abdominal distension. With regards to prevalence by occurrence, 32/78 (41%) of patients presented with two GI complications, followed by 21 patients (27%) who presented with a single GIC. Only 11 patients (14%) presented with more than three complications. Median length of stay was higher in patients with GIC (8 days) than with those who did not develop GIC (4 days). The frequency of gastrointestinal complications was significantly higher in children receiving mechanical ventilation, on sedatives and relaxants and those with multiorgan dysfunction syndrome (MODS) and inotropes. CONCLUSION: GI complications are a frequent occurrence in the PICU and are associated with worse clinical outcomes. The use of sedative drugs and the presence of shock with MODS were amongst the important contributing factors.

Spectrum of complications of severe DKA in children in pediatric Intensive Care Unit.

OBJECTIVES: To describe the spectrum of complications of Diabetic Ketoacidosis (DKA) observed in children admitted with severe DKA. METHODS: Retrospective review of the medical records of all children admitted with the diagnosis of severe DKA in Pediatric Intensive Care Unit (PICU) of the Aga Khan University Hospital, from January 2010 to December 2015 was done. Data was collected on a structured proforma and descriptive statistics were applied. RESULTS: Total 37 children were admitted with complicated DKA (1.9% of total PICU admission with 1.8% in 2010 and 3.4% in 2015). Mean age of study population was 8.1±4.6 years and 70% were females (26/37). Mean Prism III score was 9.4±6, mean GCS on presentation was 11±3.8 and mean lowest pH was 7.00±0.15. Complications observed included hyperchloremia (35.94%), hypokalemia (30.81%), hyponatremia (26.70%), cerebral edema (16.43%), shock (13.35%), acute kidney injury (10.27%), arrhythmias (3.8%), and thrombotic thrombocytopenic purpura (5.4%), while one patient had myocarditis and ARDS each. 13/37 children (35%) needed inotropic support, 11/37 (30%) required mechanical ventilation while only one patient required renal replacement therapy. Two patients (5.4%) died during their PICU stay. CONCLUSION: Hyperchloremia and other electrolyte abnormalities, cerebral edema and AKI are the most common complications of severe DKA.

Relationship of Serum Procalcitonin, C-reactive Protein, and Lactic Acid to Organ Failure and Outcome in Critically Ill Pediatric Population.

OBJECTIVE: To evaluate the clinical and prognostic utility of procalcitonin (PCT), C-reactive protein (CRP), and lactic acid in children admitted to the Pediatric Intensive Care Unit (PICU) of a university teaching hospital. MATERIALS AND METHODS: Medical records of children (1 month-16 years) tested for serum PCT at the time of admission in the PICU of our hospital from July 1, 2013, to January 15, 2015, were reviewed. Within 24 h of admission, the Pediatric Risk of Mortality Score, blood cultures, white blood cell count, neutrophil counts, serum CRP, plasma lactic acid, and PCT were noted. Patient outcome was assessed at hospital discharge, and the patients were divided into nonsurvivors and survivors. RESULTS: A total of 167 children being admitted to the PICU were enrolled. The median age of the study population was 3 years (0-16 years), with 58.6% being males. Nonsurvivors had significantly higher lactic acid (4.7 mmol/L [2.07-7.6]; P < 0.05) than that of the survivors (2 mmol/L [1.3-3]; P < 0.05). In addition, nonsurvivors (94.4%; P < 0.05) had greater incidence of multiple organ dysfunction syndrome (MODS) than that of the survivors (38.05%; P < 0.05). Binary logistic regression showed age, MODS, and lactic acid to be associated with mortality. CONCLUSIONS: This study found that in comparison to PCT and CRP, high plasma lactic acid levels are associated with the development of all-cause MODS and worse outcome in critically ill children admitted in PICU. Prediction of prognosis based on the lactic acid alone may contribute to improve patient management, but further studies are required to endorse our findings.

Hyperlactetemia And Its Trends In Critically Ill Children Admitted In Pediatric Intensive Care Unit Of A Developing Country.

BACKGROUND: There is increasing evidence that in setting of critical ailments clinical signs lag biomarkers like Lactate and hyperlactetemia can be the only marker for this disorder. This study was conducted to describe the incidence of hyperlactatemia in critically ill children and its association with outcome. METHODS: Retrospective review of medical records of all children who had their lactic acid (LA) levels measured during their admission in PICU from January 2014 to December 2015 was done. Demographic and clinical variables were recorded along with PICU therapies, outcome (Survived or expired) and development of multi-organ dysfunction. Results are presented as frequency with percentages and mean with standard deviation. Appropriate statistical tests were applied and p-value of <0.05 was taken as significant. RESULTS: Total 300 patients had their LA measured and 202 were included in the study. Males were 130 (64%) and mean age was 5.7±4.6 years. Hyperlactatemia was found in 68 (33%) patients and another 75 (37%) had a second LA level >4 mmol/L. Increasing LA trend was found in 79 (39%) patients. Diagnostic categories included cardiovascular diseases (45, 22%), central nervous system diseases (40, 20%), respiratory diseases (31, 15%), sepsis (28, 14%), and gastrointestinal diseases14 (7%). 168 (83%) needed mechanical ventilation. Mean pH was 7.31±0.15 and metabolic acidosis was observed in 91 patients (45%). Mean LA levels in survivors and non survivors were 3.3±3.12 and 5.35±5.47 respectively. Hyperlactatemia was associated with death (p=0.01) and development of MODS (p=0.03) on univariate analysis. On multivariate logistic regression rising lactate and development of MODS were significantly associated with death (p=<0.05, odds ratio (OR) 9.24 (95% confidence interval 1.55-55.20). CONCLUSIONS: Hyperlactatemia and increasing LA trend in critically ill children are associated with worse outcome in PICU.

Pneumonia classification: A limited data approach for global understanding.

As the human race has advanced, so too have the ailments that afflict it. Diseases such as pneumonia, once considered to be basic flu or allergies, have evolved into more severe forms, including SARs and COVID-19, presenting significant risks to people worldwide. In our study, we focused on categorizing pneumonia-related inflammation in chest X-rays (CXR) using a relatively small dataset. Our approach was to encompass a comprehensive view, addressing every potential area of inflammation in the CXR. We employed enhanced class activation maps (mCAM) to meet the clinical criteria for classification rationale. Our model incorporates capsule network clusters (CNsC), which aids in learning different aspects such as geometry, orientation, and position of the inflammation seen in the CXR. Our Capsule Network Clusters (CNsC) rapidly interpret various perspectives in a single CXR without needing image augmentation, a common necessity in existing detection models. This approach significantly cuts down on training and evaluation durations. We conducted thorough testing using the RSNA pneumonia dataset of CXR images, achieving accuracy and recall rates as high as 98.3% and 99.5% in our conclusive tests. Additionally, we observed encouraging outcomes when applying our trained model to standard X-ray images obtained from medical clinics.

Neutrophil Gelatinase-Associated Lipocalin as a Predictor of Acute Kidney Injury in Children With Shock: A Prospective Study.

BACKGROUND: The current definition of acute kidney injury (AKI) is based on serum creatinine (SrCr) and urine output, limited by delayed identification of such patients. Plasma neutrophil gelatinase-associated lipocalin (NGAL) is considered an early diagnostic and highly predictive biomarker of AKI. OBJECTIVE: To determine the diagnostic accuracy of NGAL for AKI compared with creatinine clearance for early detection of AKI in children with shock receiving inotropic support. METHODS: Critically ill children requiring inotropic support in the pediatric intensive care unit were enrolled prospectively. SrCr and NGAL values were obtained three times at six, 12, and 48 hours after vasopressor initiation. Patients with AKI were defined as having loss of >25% renal function based on creatinine clearance within 48 hours. NGAL level of more than 150 ng/dl was suggestive of the diagnosis of AKI. Receiver operator characteristic curves were generated for NGAL and SrCr to compare the predictive ability of both at 0, 12, and 48 hours of starting vasopressor support.  Results: A total of 94 patients were enrolled. The mean age was 43±50.95 months. Most common primary diagnoses were related to the cardiovascular system (46%). Twenty-nine patients (31%) died during the hospital stay. Thirty-four patients (36%) developed AKI within 48 hours following shock. The area under the curve (AUC) for NGAL at a cutoff of 150 ng/ml was 0.70, 0.74, and 0.73 at six-hour, 12-hour, and 48-hour follow-up, respectively. NGAL had a sensitivity of 85.3% and specificity of 50% at 0 hours of follow-up for diagnosis of AKI. CONCLUSION: Serum NGAL has better sensitivity and AUC compared to SrCr for early diagnosis of AKI in children admitted with shock.

Comparison Of Two Definitions (P-Rifle And Kdigo) For Prevalence Of Acute Kidney Injury And In Hospital Mortality In A Paediatric Intensive Care Unit Of Pakistan.

BACKGROUND: To compare the Paediatric RIFLE (p-RIFLE) and Kidney Disease Improving Global Outcomes (KDIGO) definitions of acute kidney injury (AKI) for frequency of (AKI) and in-hospital mortality in critically ill children. METHODS: Retrospective review of medical records of all patients (aged 1 month - 16 years) admitted in Paediatric Intensive Care Unit from January 2015-December 2016, with length of stay >48 hours, was done. Patients with chronic kidney disease were excluded. Receiver operating characteristic (ROC) curves were used to evaluate the performance of the p-RIFLE and KDIGO criteria to predict the AKI related mortality. Logistic regression analysis was done to determine the association of different variables with mortality in AKI patient based on p-RIFLE, KDIGO. A p-value of <0.05 was considered significant. RESULTS: Out of total 823 patients admitted during the study period, 562 patients were included in the study. Median age was 2 years (Interquartile range 8 years). Acute kidney injury frequency according to p-RIFLE and KDIGO were 391 (70%), and 372 (66%) respectively. Overall, 106/823 (12.8%) children died during study period, 78 (19.9%) in AKI by p-RIFLE and 76 (20.4%) in AKI by KDIGO died. The area-under- curve for in-hospital mortality for p-RIFLE and KDIGO criteria were 0.525 (p=0.427), and 0.534 (p=0.276), respectively. CONCLUSIONS: P-RIFLE is more sensitive compared to KDIGO in diagnosing AKI in critically ill children; identifying a greater number of moderate staged AKI cases. Greater AKI severity is associated with higher mortality in critically ill children.

Early postoperative arrhythmias after pediatric congenital heart disease surgery: a 5-year audit from a lower- to middle-income country.

BACKGROUND: Arrhythmias are known complication after surgery for congenital heart disease (CHD). This study aimed to identify and discuss their immediate prevalence, diagnosis and management at a tertiary care hospital in Pakistan. METHODS: A retrospective study was conducted at a tertiary care hospital in Pakistan between January 2014 and December 2018. All pediatric (<18 years old) patients admitted to the intensive care unit and undergoing continuous electrocardiographic monitoring after surgery for CHD were included in this study. Data pertaining to the incidence, diagnosis, and management of postoperative arrhythmias were collected. RESULTS: Amongst 812 children who underwent surgery for CHD, 185 (22.8%) developed arrhythmias. Junctional ectopic tachycardia (JET) was the most common arrhythmia, observed in 120 patients (64.9%), followed by complete heart block (CHB) in 33 patients (17.8%). The highest incidence of early postoperative arrhythmia was seen in patients with atrioventricular septal defects (64.3%) and transposition of the great arteries (36.4%). Patients were managed according to the Pediatric Advanced Life Support guidelines. JET resolved successfully within 24 hours in 92% of patients, while 16 (48%) patients with CHB required a permanent pacemaker. CONCLUSIONS: More than one in five pediatric patients suffered from early postoperative arrhythmias in our setting. Further research exploring predictive factors and the development of better management protocols of patients with CHB are essential for reducing the morbidity and mortality associated with postoperative arrhythmia.

Frequencies of different nuclear morphological features in prostate adenocarcinoma.

Diagnosis of prostate adenocarcinoma is primarily based on morphological analysis. Nucleomegaly, prominent nucleoli, and hyperchromasia constitute current nuclear diagnostic parameters but are seen in benign conditions, vary with Gleason grade, and pose diagnostic challenge in well-differentiated tumors with accompanying inflammation or equivocal architectural features. In such cases, other pleomorphic nuclear features such as variation in size and shape, irregular contours, nuclear membrane infoldings, and nonuniform chromatin, which are not incorporated in formal evaluation, may prove helpful. Our aim was to study different nuclear morphological features of prostate adenocarcinoma (including currently practiced ones) and highlight their variation with Gleason grades. We examined 84 cases of prostate adenocarcinoma using oil immersion microscopy where necessary. Commonest Gleason pattern observed was grade 4a accounting for 42.8% of cases. Nuclear enlargement (moderate to marked in 93.8%), nucleolar enlargement (62.1%), and nonuniform chromatin distribution (100%) could serve as useful diagnostic features but did not vary with tumor differentiation. Pleomorphism (moderate in 58.6%), nuclear overlapping (62.8%), nuclear membrane infoldings (66.2%), and irregular contours (frequent in 94.5%) were significant diagnostic features that increased in frequency and extent with increasing grade and could be used to differentiate low-grade from high-grade tumors. Worsening of nuclear morphology with advancing tumor grades indicated that nuclear anaplasia accompanies poor architectural differentiation. Coexistence of pale and dark nuclei signified variable chromatin density of no diagnostic significance.

Metaplastic squamous cell carcinoma in association with invasive lobular breast carcinoma with metastasis to axillary lymph nodes.

Squamous cell carcinoma (SCC) of the breast is very rare entity and when encountered it is found in association with invasive ductal carcinoma of the breast. We report a very unusual association of an invasive lobular carcinoma with metaplastic squamous cell carcinoma of the breast which was found not only in a mastectomy specimen but also in the axillary lymph nodes. The patient reported first with a lump breast which was diagnosed a case of invasive lobular carcinoma in a lumpectomy specimen, followed by adjunct radiation and chemotherapy. We encountered such association in only one case report in the literature.

High Flow Nasal Cannula Therapy in Children with Acute Respiratory Insufficiency in the Pediatric Intensive Care Unit of a Resource-limited Country: A Preliminary Experience.

A retrospective study was performed in children (aged one month - 16 years) receiving HFNC to determine the frequency, efficacy and adverse effects of high flow nasal cannula (HFNC) therapy in the pediatric intensive care unit (PICU), from January to December 2017. Treatment failure was defined as clinical deterioration on HFNC therapy such that mechanical ventilation (MV) was required. Clinical parameters before and after HFNC were assessed using repeated measures analysis of variance. A total of 120 patients received HFNC therapy (21% of total admissions). Primary diagnosis were respiratory disease (50%), central nervous system diseases (14.2%), sepsis (10.8%), and postoperative care (10%). Mean duration of HFNC was 27.5 ±19.7 hours and mean PICU length of stay was 6 ± 6 days. Pneumothorax developed in four patients. MV was required in 28 patients, and subsequently, 15 deaths occurred in that group. HFNC is a frequently used, safe and effective therapy for children requiring respiratory support in PICU. Key Words: Respiratory insufficiency, Non-invasive ventilation, Paediatric intensive care unit.

Safety and Efficacy of Aminophylline in Intrathecal Methotrexate-related Neurological Toxicity in Large Pediatric Oncology Centre.

Children with acute lymphoblastic leukemia, receiving intrathecal methotrexate (IT-MTX), develop acute and subacute neurological dysfunction in 3-15% of cases. Altered level of consciousness, seizure, and stroke-like manifestations are among the most common presentations. MRI of the brain in an early stage is consistent with white matter leukoencephalopathy. There are no specific guidelines for the treatment of such a syndrome. An elevated cerebrospinal fluid (csf) adenosine concentration causes vasodilatation in the brain and leads to cerebral ischemia, which may be reduced by aminophylline infusion. The study is a retrospective data analysis in which electronic data records of 30 patients, collected from September 2017 - August 2019. The primary objective of the study was to evaluate the safety and efficacy of aminophylline in IT-MTX induced neurotoxicity. Aminophylline infusion was used in 30 patients with IT-MTX neurotoxicity. Twenty-five patients (83.33%) showed dramatic improvement of neurologic signs and symptoms within 48 hours. Aminophylline was, hence, coined as a salvageable therapy. No noticeable side effects were observed during treatment with aminophylline infusion. Key Words: Intrathecal methotrexate, Neurotoxicity, Children, Cancer, Aminophylline.

Efficacy and Safety of Phosphate Enema as a therapeutic agent in Hypophosphatemia in Critically Ill Children.

The aim of the study was to evaluate the safety and efficacy of oral administration of phosphorous enema in hypophosphatemia (HP) in critically ill children admitted in the pediatric intensive care unit (PICU) of The Indus Hospital, Karachi, from September 2018 to August 2019. This was a retrospective review of 31 critically ill children with hypophosphatemia who received 1 ml/kg/day of phosphate enema through nasogastric tube or orally for phosphate replacement, with serial phosphorus level monitoring along with observation for its side effects. The results showed that the rise of serum phosphorus level was observed in all cases and 64.5% of cases achieved target phosphorus level with no adverse reactions observed. Sample size although limited, it is safe to state that oral phosphate enema is safe and effective for correction of hypophosphatemia in critically ill children. Key Words: Hypophosphatemia, Enema, Pediatric intensive care unit.