Risk factors for ischemia/necrosis of the colonic stump after proctectomy and delayed coloanal anastomosis.

INTRODUCTION: Delayed coloanal anastomosis (DCAA) is a two-stage procedure. DCAA has been increasingly reused in recent years in the management of rectal cancer. Such increased use of DCAA has highlighted the complications associated with this procedure. We aimed to evaluate the risk and risk factors of ischemia/necrosis of the colonic stump between the two stages of DCAA. PATIENTS AND METHODS: All patients who underwent a proctectomy with a DCAA were included in this retrospective single-centre study from November 2012 to June 2022. Two groups of patients were defined: patients with a well vascularized colonic stump (well vascularized group) and those who experienced ischemia or necrosis of the colonic stump (ischemic group). The primary endpoint was the rate of ischemia or necrosis of the colonic stump and an evaluation of the associated risk factors. RESULTS: During the study period, 43 patients underwent DCAA. Amongst them, 32 (75%) had a well-vascularized colonic stump (well-vascularized group) and 11 (25%) ischemia of the colonic stump (ischemic group). Relative to patients in the well-vascularized group, those in the ischemic group were more often men (81.8% vs 40.6%, p = 0.034), had a higher BMI (29.2 kg/m2 vs 25.7 kg/m2, p = 0.03), were more frequently diabetic (63.6% vs 21.9%, p = 0.01) and more frequently had had preoperative radiotherapy (100% vs 53.1%, p = 0.008). On the preoperative CT scan, the interspinous diameter was shorter in the ischemic group (9.4 ± 1.01 cm vs 10.6 ± 1.01 cm, p = 0.001), the intertuberosity diameter was shorter (9.2 ± 1.18 cm vs 11.9 ± 1.18 cm, p < 0.0001), and the length of the anal canal was longer (31.9 ± 3 mm vs 27.4 ± 3.2 mm, p < 0.0001). CONCLUSION: This study highlights clear risk factors for ischemia/necrosis of the colonic stump after proctectomy with DCAA.

Clinical impact of microbiological results in two-stage revision arthroplasty with spacer exchange.

INTRODUCTION: Patients who require a spacer exchange as part of a two-stage procedure for the treatment of periprosthetic hip and knee joint infections (PJI) have high failure rates. Little is known about the clinical impact of microbiological results and changes in the microbiological spectrum and resistance pattern in these patients. MATERIAL AND METHODS: Between 01/2011 and 12/2019, 312 patients underwent a total of 327 two-stage revision arthroplasties at our institution. A spacer exchange was required in 52/312 (16.7%) patients (27 knee/25 hip). Microbiological results, antibiotic resistance patterns, patient's host factors as well as re-revision and re-infection rates at a median follow-up of 47.8 months (range 12.2-116.7 months) were analyzed. A propensity score (PS)-matched analysis of patients who underwent spacer exchange and patients treated with standard two-stage procedure was performed. RESULTS: We found a high number of microbiological spectrum changes in patients with multiple culture positive procedures between explantations and spacer exchanges (10/12 [83.3%]), spacer exchanges and reimplantations (3/4 [75%]) as well as between reimplantations and subsequent re-revision surgeries (5/6 [83.3%]). In 9/52 (17.3%) patients, same microorganisms were detected repeatedly in two different procedures. We observed changes in the antibiotic resistance patterns in 6/9 (66.7%) of these patients. High re-infection rates were found in patients with culture positive reimplantations (10/12 [83.3%]), and low re-infection rates were found in patients with culture negative reimplantations (2/40 [5%]; p < 0.001). Between patients with and without spacer exchange, no differences were found in the re-revision rates (13/52 [25%] with vs. 13/52 [25%] without; p = 1.00) as well as re-infection rates (12/52 [23.1%] with vs. 8/52 [15.4%] without; p = 0.32). CONCLUSIONS: Changes in microbiological spectrum and antibiotic resistance patterns between stages are common in patients who require a spacer exchange. If eradication of the microorganism at reimplantation can be accomplished, comparable re-revision rates to standard two-stage procedures can be achieved.

On Fixed Accuracy Confidence Interval in Multivariate Normal Distribution with Order 1 Autoregressive Covariance Structure.

In this paper, stein-type two-stage sampling procedure is carried out for fixed accuracy confidence interval estimation of the common variance ( σ 2 ) parameter corresponding to multivariate normal distribution with autoregressive covariance structure of order 1. Related asymptotics are obtained and simulation results are presented.

Is the Two-Stage Construction Approach Better for Maturation and Long-term Patency of the Transposed Brachiobasilic Arteriovenous Fistulas?

The transposed brachiobasilic arteriovenous fistula is often constructed by employing a two-stage procedure for better maturation and long-term patency, but its validity on overall outcomes remains controversial. A single-center's experience is presented herein, comparing one- and two-stage approaches on 229 consecutive first-time arteriovenous access constructed in the upper arm with a loop configuration and an obtuse anastomosis angle that provide favorable flow dynamics. The one- and two-year patency rates were equivalent between the two approaches, but the fistula rate was greater and the maturation time was shorter with the one-stage procedure.

One-stage versus two-stage Roux-Y gastric bypass as redo surgery of failed adjustable gastric banding.

INTRODUCTION: The study focussed on whether a one-stage Roux-Y gastric bypass (OS-RYGB) or a two-stage RYGB (TS-RYGB) has a significant advantage in terms of perioperative risk in patients after failed adjustable gastric banding (AGB). METHODS: Data collection included patients who underwent OS-RYGB or TS-RYGB after AGB between 2005 and 2019 and whose outcomes were compared with those after primary RYGB (P-RYGB). Outcome criteria were perioperative complications, comorbidities, 30-day mortality and operating time. RESULTS: The study analysed data from patients who underwent OS-RYGB (N = 525), TS-RYGB (N = 382) and P-RYGB (N = 26,445). Intraoperative and postoperative complication rates were significantly lower for P-RYGB (p < 0.001). Total intraoperative and specific postoperative complication rates were significantly lower in TS-RYGB than in OS-RYGB (p = 0.048 and p < 0.001, respectively). In contrast, the total general postoperative complication rate was lower in OS-RYGB than in TS-RYGB (p < 0.001). The mean operating time differed significantly among the three groups (P-RYGB 96.5min, OS-RYGB 141.2min and TS-RYGB 190.9min; p < 0.001). The mortality rate was not significantly different between the three groups. CONCLUSIONS: Based on the significant difference between the two groups in revision surgery and the slight difference with the results of primary RYGB, this study concludes that removal of a failed AGB is safe and feasible with either the OS- or TS-RYGB procedure. However, we cannot directly recommend either procedure in our study. Proper patient selection and surgeon experience are critical to avoid potential adverse effects.

Analysis of Failed Two-Stage Procedures with Resection Arthroplasty as the First Stage in Periprosthetic Hip Joint Infections.

Resection arthroplasty can be performed as the first stage of a two-stage procedure in some patients with severe periprosthetic hip joint infections with poor bone stock. This retrospective study aimed to evaluate factors associated with the subsequent failure or success of these patients. Between 2011 and 2020; in 61 (26.4%) of 231 patients who underwent a two-stage protocol of periprosthetic hip joint infections; no spacer was used in the first stage. The minimum follow-up period was 12 months. Patient's demographics and various infection risk factors were analyzed. In total, 37/61 (60.7%) patients underwent a successful reimplantation, and four patients died within the follow-up period. Patients within the failure group had a significantly higher Charlson comorbidity index (p = 0.002); number of operations prior to resection arthroplasty (p = 0.022) and were older (p = 0.018). Failure was also associated with the presence of a positive culture in the first- and second-stage procedures (p = 0.012). Additional risk factors were persistent high postoperative CRP values and the requirement of a negative-pressure wound therapy (p ≤ 0.05). In conclusion, multiple factors need to be evaluated when trying to predict the outcome of patients undergoing resection arthroplasty as the first stage of a two-stage procedure in patients with challenging periprosthetic hip joint infections.

Comparative study of one-stage and the novel two-stage urethroplasty using the transected urethral plate and transverse preputial island flap (TPIF) for severe hypospadias: a single-center experience.

BACKGROUND: To evaluate a novel two-stage method of urethroplasty using a transected urethral plate and transverse preputial island flap (TPIF) for proximal hypospadias with severe chordee. METHODS: We retrospectively analyzed 91 cases of proximal hypospadias, who had undergone either one-stage or two-stage urethroplasty. Comparisons between the two groups were made based on age, glans diameter, the length of the urethral defect, and the rate of complications. RESULTS: There was no significant difference in age at surgery, glans diameter, and the length of urethral defect between the two groups. In the one-stage group, 18 cases of urinary fistula and one case each of glans dehiscence, urethral diverticulum, and urethral diverticulum with concurrent glans dehiscence, were reported. All patients were treated using urinary fistula repair or urethroplasty. In the novel two-stage group, no cases of urinary fistula were reported after the secondary urethroplasty. However, 4 cases showed dehiscence at the glans, 2 were treated through reoperation but the 2 other 2 patients reoperation. The success rate was 62.50% and 88.57% in the one-stage and two-stage groups, respectively (P<0.05). CONCLUSIONS: Two-stage urethroplasty using a tubularized urethral plate and TPIF can significantly reduce the incidence of urethral fistula of the proximal hypospadias with severe chordee. However, it may increase the number of operations that need to be performed on children who do not need staged surgery. Although we could not prove it through this study, we believe that the necessity of two-stage urethroplasty should be determined based on the development of the glans, the severity of penile curvature, and the length of the urethra defect.

Staged Bilateral Total Knee Arthroplasty: When Should the Second Knee be Replaced?

BACKGROUND: Bilateral total knee arthroplasty (BTKA) under the same anesthesia (simultaneous) or staged are options for patients with end-stage arthritis of the knee that carries advantages and limitations. Not all patients are candidates for simultaneous BTKA, and therefore, surgeons prefer to stage the two TKAs. The optimal safe interval between two TKAs is not known. The present systematic review aimed to determine the optimal time interval between the two stages of BTKA. METHODS: Pubmed and Scopus databases were searched to identify publications from January 1979 to November 2017 in English that compared the outcomes of staged BTKA performed using various time intervals between the two TKAs. Data on systemic and local complications following staged BTKA were extracted, and the pooled data were analyzed to adjust for age. RESULTS: In total, 23 studies that enrolled 117,090 patients undergoing staged BTKA were included in this systematic review. A significant increase was observed in the incidence of myocardial infarction (OR=8.4 and 8.32), other cardiac complications (OR=17.71 and 18.18), deep vein thrombosis (OR=4.72 and 4.89), pneumonia (OR=3.37 and 3.45), and knee revision (OR=3.73 and 4.14) in patients undergoing the second TKA within 30 days or 90 days of the first TKA. However, the replacement of the second knee within this time interval was associated with a significantly lower risk of pulmonary embolism (OR=0.145 and 0.128), superficial (OR=0.14 and 0.79) and deep knee infection (OR=0.0 and 0.0), as well as vascular complications (OR=0.0 and 0.42). CONCLUSION: Time interval of less than 30 or 90 days between two TKAs performed in patients with BTKA was associated with a higher risk of systematic complications. However, the shorter time intervals between the two TKA may reduce the risk of other complications. This information may help surgeons' council patients better when deciding on the optimal time interval between two TKAs.

A two-stage approach to primary TKA using articulating antibiotic-loaded spacers improve function and eradicate infection in septic arthritic knees.

PURPOSE: The treatment of an infected arthritic knee might be challenging. The failure rate has been reported to be high for open or arthroscopic debridement. A subsequently high rate of infection has been noted in these patients undergoing primary total knee arthroplasty (TKA). In the present study, a two-stage approach using an articulating spacer was used. The hypothesis was that the procedure would eradicate the infection and improve pain and function in these patients. METHODS: A total of 16 consecutive patients were enrolled in this retrospective study. The mean follow-up time was 6.1 years (range 2.0-9.9 years). Patients with advanced osteoarthritis and infection of the knee were included. All patients had previously undergone one or more failed arthroscopic or open procedures for the eradication of infection. All patients received the same homemade metal-on-plastic articulating antibiotic spacer. Double antibiotic therapy was given for 2 weeks intravenously and orally for 4 weeks. TKA implantation was performed 6 weeks after the first stage. RESULTS: The infection was eradicated without recurrence in all patients. The functional results were significantly improved, and pain was significantly reduced after spacer and TKA implantation. The mean amount of knee flexion was 95 ± 30° preoperatively, and it increased to 109 ± 14° (p = 0.012) after spacer implantation and to 119 ± 10° (p = 0.002) after TKA implantation. The mean KSS objective was 58 ± 12 preoperatively, and it increased to 75 ± 14 (p < 0.0001) after spacer implantation and to 96 ± 3 (p < 0.0001) after TKA implantation. The mean KSS function was 17 ± 11 preoperatively, and it increased to 46 ± 10 (p < 0.0001) after spacer implantation and to 86 ± 6 (p < 0.0001) after TKA implantation. The mean VAS score was 65 ± 11 preoperatively, and it decreased to 2 ± 4 (p < 0.0001) after spacer implantation and to 1 ± 2 (p < 0.0001) after TKA implantation. CONCLUSION: The two-stage procedure for the treatment of infected arthritic knees after failed eradication surgery was effective in all patients. Using an antibiotic articulating metal-on-plastic cement spacer showed improved functional results between the stages and at the final follow-up. No intra- or postoperative complications occurred.

A new sequential animal model for infection-related non-unions with segmental bone defect.

BACKGROUND: The treatment of fracture-related infections (FRI) is still a challenge for orthopedic surgeons. The prevalence of FRI is particularly high in open fractures with extensive soft-tissue damage. This study aimed to develop a new two-step animal model for non-unions with segmental bone defects, which could be used to evaluate new innovative bone substitutes to improve the therapeutic options in humans with FRI and bone defects. METHODS: After randomization to infected or non-infected groups, 30 Sprague-Dawley rats underwent a transverse osteotomy of the mid-shaft femur with a 5 mm defect. Additionally, the periosteum at the fracture zone was cauterized at both sides. After intramedullary inoculation with 103 CFU Staphylococcus aureus (infected group) or PBS (non-infected group), a fracture stabilization was done by intramedullary K-wires. After 5 weeks, the bone healing process was evaluated, and revision surgery was performed in order to obtain increased bone healing. The initial K-wires were removed, and debridement of the osteotomy-gap was done followed by a more stable re-osteosynthesis with an angle-stable plate. After further 8 weeks all rats were euthanized and the bone consolidation was tested biomechanically and the callus formation quantitatively by micro-CT analysis. RESULTS: We developed and presented a new two-stage non-union animal model through a targeted S. aureus infection. After 5 weeks, all animals showed a non-union irrespective of assignment to the infected and non-infected group. Lane and Sandhu score showed a higher callus formation in the infected group. In all infected animals, the inoculated S. aureus strain was detected in the revision surgery. The second surgery did not improve bone healing, as shown by the Lane Sandhu score and in the µ-CT analysis. Similarly, biomechanical testing showed in both groups a significantly lower maximum torque as compared to the contralateral side (p < 0.0001). CONCLUSIONS: We were able to successfully develop a new two-stage non-union animal model, which reflects a genuine clinical situation of an infection-related non-union model with segmental bone defects. This model could be used to evaluate various therapeutic anti-infectious and osteoinductive strategies in FRIs.

Pediatric ulcerative colitis: three- versus two-stage colectomy with ileal pouch-anal anastomosis.

BACKGROUND: Despite advancements in medical therapy for ulcerative colitis (UC), a significant proportion of children progress to colectomy with ileal pouch-anal anastomosis (IPAA). Procedural related complications between two- and three-stage operations in children have not been well described. METHODS: We performed a retrospective review of patients who underwent a colectomy for UC or inflammatory bowel disease unclassified between 2008 and 2018. RESULTS: Forty-nine children underwent an IPAA at the time of colectomy (two stage) or during a subsequent operation (three stage). Preoperative hemoglobin and albumin concentrations were lower among those undergoing three-stage procedures. The rate of early complications (≤30 days) was similar between the two groups (p = 0.46); however, late complications (>30 days) were more commonly associated with three-stage procedures (p = 0.03). Time with a stoma was 3.2 months longer among those who underwent a three-stage procedure. While three-stage procedures were more often performed during the first half of the study period (2008-2012), two-stage procedures became more common during the second half (2013-2018). During this transition to favor two-stage procedures, complication rates did not significantly change. CONCLUSION: Although three-stage procedures were thought to be associated with fewer complications, we found comparable complication rates as we transition to two-stage procedures.

Prevalence and Predictors of Unnecessary Endoscopic Retrograde Cholangiopancreatography in the Two-Stage Endoscopic Stone Extraction Followed by Laparoscopic Cholecystectomy.

BACKGROUNDS: A two-stage procedure involving endoscopic retrograde cholangiopancreatography (ERCP), followed by cholecystectomy, is one of the primary treatments of concomitant gallstones and choledocholithiasis. However, negative findings on ERCP and migrating gallstones after cholecystectomy are major concerns. This study aimed to identify the prevalence of unnecessary ERCP and to develop and validate a predictive nomogram using preoperative factors in patients who underwent a two-stage procedure. METHODS: Consecutive 931 patients were treated with the two-stage procedure for evident gallstones and suspected choledocholithiasis. After the cholecystectomy, a cholangiogram was performed to confirm the absence of the migrating gallstones. The patients were divided into derivation (n = 652) and validation (n = 279) cohorts. RESULTS: A total of 26.5% (247/931) patients had unnecessary ERCP (negative choledocholithiasis, 14.6%; migrating gallstones, 11.9%). No stones on images (P < 0.001), total bilirubin < 1.2 mg/dL (P = 0.006), and common bile duct diameter < 8.0 mm (P = 0.004) were independent factors associated with negative finding on ERCP with a validated nomogram area under the curve (AUC) of 0.72 (95% confidence interval [CI] 0.64-0.80). For migrating gallstones after cholecystectomy, radiolucent gallstones (P < 0.001), gallstone size ≤ 6.4 mm (P = 0.001), cystic duct stones (P < 0.001), gallbladder wall thickness ≥ 3.2 mm (P = 0.003), and low-lying cystic duct (P < 0.001) were independent factors with a validated nomogram AUC of 0.77 (95% CI 0.68-0.87). CONCLUSIONS: About one fourth of the patients may have unnecessary ERCP in the two-stage procedure. Based on our nomogram using preoperative factors, high-risk patients who are more likely to perform unnecessary ERCP could be considered for the one-stage procedure.

Laparoscopic two-stage procedure for gallstone ileus.

Gallstone is a rare cause of bowel obstruction and is associated with high rates of morbidity and mortality. Here, we report a case of gallstone ileus treated by laparoscopic two-stage procedure. A 65-year-old man, without a history of any surgery, presented to our hospital with abdominal pain and vomiting. His abdominal computed tomography revealed gallstone ileus with cholecystoduodenal fistula. Then, enterolithotomy was performed as an emergency laparoscopic surgery. After 1 year of the surgery, a second laparoscopic procedure was performed for cholecystectomy and fistula closure. The patient was discharged on the 7th postoperative day. Laparoscopic two-stage procedure is a safe and an efficient approach for the management of gallstone ileus. This definitive treatment is minimally invasive, thereby suitable for treating elderly patients.

Comparison of two-stage revision arthroplasty and intramedullary arthrodesis in patients with failed infected knee arthroplasty.

BACKGROUND: Failed total knee arthroplasty (TKA) with significant bone loss and compromised soft-tissues is challenging and the final results are often inferior to patient's expectation. The objective of this study was to present a comparison of outcomes in patients with failed infected TKA treated with two-stage revision TKA or knee arthrodesis and to assess clinical and functional results, implant survival and infection recurrence. The hypothesis was that an arthrodesis may result in beneficial effects on patients' outcome. METHODS: Clinical data of 81 patients with periprosthetic joint infection (PJI) of the knee joint were collected and analyzed retrospectively. Between 2008 and 2014, a total of 36 patients had been treated within a two-stage exchange procedure and reimplantation of a modular intramedullary arthodesis nail and 45 patients with revision TKA. Patients were treated according to the same structured treatment algorithm. Clinical and functional evaluation was performed using the Oxford knee score (OKS) and the visual analogue scale (VAS). RESULTS: The mean follow-up was 32.9 ± 14.0 months. The rate of definitely free of infection at last follow-up in the arthrodesis group was 32 of 36 (88.9%) and 36 of 45 (80.0%) in the revision TKA group (p = 0.272). Mean VAS for pain in the arthrodesis group was 3.1 ± 1.4 compared to 3.2 ± 1.6 in the revision TKA group (p = 0.636). The OKS in the arthrodesis group was 38.7 ± 8.9 and 36.5 ± 8.9 (p = 0.246) in patients with revision TKA. Rate of revisions in the revision-TKA group was 2.8 ± 3.7 compared to 1.2 ± 2.4 in the arthrodesis group (p = 0.021). CONCLUSION: Treatment of PJI needs a distinct therapy with possible fallback strategies in case of failure. A knee arthrodesis is a limb salvage procedure that showed no significant benefits on the considered outcome factors compared to revision TKA but is associated with significantly lower revision rate. After exhausted treatment modalities, a knee arthrodesis should be considered as an option in selected patients. LEVEL OF EVIDENCE: Level III, retrospective cohort study.

Two-stage knee arthrodesis with a modular intramedullary nail due to septic failure of revision total knee arthroplasty with extensor mechanism deficiency.

BACKGROUND: Periprosthetic joint infection is a serious complication and reconstruction after failed revision total knee arthroplasty with significant bone loss and compromised soft-tissues can be challenging. Objective of this study was to assess clinical and functional results, implant survival and infection recurrence rates in patients treated with two-stage arthrodesis after failed revision TKA with extensor mechanism deficiencies due to PJI, and to identify the factors that affect outcomes after surgery. METHODS: Thirty seven patients with PJI treated within a two-stage exchange and reimplantation of an arthrodesis nail between 2008 and 2014 were included. Systemic and local risk factors were graded preoperatively according to McPherson et al. All patients were treated according to a structured treatment algorithm. Clinical and functional evaluation was performed using the Oxford Knee Score and the Visual Analogue Scale. RESULTS: Thirty two of 37 patients (86.5%) were graded as free of infection. Five patients (13.5%) had recurrent infection after arthrodesis with the need of revision surgery. Mean leg-length discrepancy was 2.2cm. The mean VAS for pain was three, the mean Oxford Knee Score was 38±9. Total implant survival at a 74month follow-up was 74.3% (95% CI: 45.4 to 91.1%), as determined by Kaplan-Meier survival curves. Local McPherson Score, as well as number of revisions was found to be of significant influence to the survival rate. CONCLUSIONS: Septic failure of revision knee arthroplasty can be effectively treated with two-stage arthrodesis using a modular intramedullary nail, providing a stable and painless limb with satisfactory functional results and acceptable infection eradication rates.

Brachial Artery-Brachial Vein Fistula for Hemodialysis: One- or Two-Stage Procedure-A Review.

Purpose The purpose of the study was to investigate the patency rate of the brachial vein transposition-arteriovenous fistula [BVT-AVF] and to review the available literature regarding the comparison of the one-stage with the two-stage procedure. Methods A multiple electronic health database search was performed, aiming to identify studies on brachial vein superficialization. Case reports and series with five or less patients were excluded from the study. End points of the study were the patency rates at 12, 24, and 36 months of follow-up. The patency of the one- or two-stage BVT-AVF procedure was investigated. Results Overall, 380 BVT-AVFs were analyzed. The primary patency rate at 12 months ranged between 24 and 77%. Rate of early fistula malfunction or failure of maturation of the fistula resulting in loss of functionality ranged from 0 to 53%. Forearm edema, hematomas, wound infection, and early thrombosis were among the most common complications. Limited data were available for the comparison of patency rates between the one- and the two-stage procedure because of the absence of sufficient comparative studies. However, series with one-stage procedure presented a lower patency rate at 12 months compared to series with two stages. Conclusion Patency rates after BVT-AVF, although not excellent, has encouraging results taking into account that patients undergoing these procedures do not have an accessible superficial vein network; failure of maturation and the increased rate of early postoperative complications remain a concern. The BVT-AVF is a valuable option for creating an autologous vascular access in patients lacking adequate superficial veins.

Reader reaction: A note on the evaluation of group testing algorithms in the presence of misclassification.

In the context of group testing screening, McMahan, Tebbs, and Bilder (2012, Biometrics 68, 287-296) proposed a two-stage procedure in a heterogenous population in the presence of misclassification. In earlier work published in Biometrics, Kim, Hudgens, Dreyfuss, Westreich, and Pilcher (2007, Biometrics 63, 1152-1162) also proposed group testing algorithms in a homogeneous population with misclassification. In both cases, the authors evaluated performance of the algorithms based on the expected number of tests per person, with the optimal design being defined by minimizing this quantity. The purpose of this article is to show that although the expected number of tests per person is an appropriate evaluation criteria for group testing when there is no misclassification, it may be problematic when there is misclassification. Specifically, a valid criterion needs to take into account the amount of correct classification and not just the number of tests. We propose, a more suitable objective function that accounts for not only the expected number of tests, but also the expected number of correct classifications. We then show how using this objective function that accounts for correct classification is important for design when considering group testing under misclassification. We also present novel analytical results which characterize the optimal Dorfman (1943) design under the misclassification.

Penalized Q-Learning for Dynamic Treatment Regimens.

A dynamic treatment regimen incorporates both accrued information and long-term effects of treatment from specially designed clinical trials. As these trials become more and more popular in conjunction with longitudinal data from clinical studies, the development of statistical inference for optimal dynamic treatment regimens is a high priority. In this paper, we propose a new machine learning framework called penalized Q-learning, under which valid statistical inference is established. We also propose a new statistical procedure: individual selection and corresponding methods for incorporating individual selection within penalized Q-learning. Extensive numerical studies are presented which compare the proposed methods with existing methods, under a variety of scenarios, and demonstrate that the proposed approach is both inferentially and computationally superior. It is illustrated with a depression clinical trial study.