AbstractObjectiveTo investigate whether physician–patient agreement of potential patient problem areas impacts subsequent patient enrollment in an interdisciplinary pain management program.DesignRetrospective chart review of 544 patients who underwent evaluation of their chronic pain. Physicians and their patients endorsed perceived patient problems during the evaluation. The potential problems included 7 clinical domains: pain, sleep, mood, physical functioning, ability to cope with pain, ability to manage pain flare‐ups, and pain medication effectiveness.ResultsResults indicated statistically significant levels of agreement among the physicians and their patients (free‐marginal kappa range, 0.19 to 0.94, P's < 0.001). The highest agreement occurred for pain and the lowest for pain medication effectiveness. Patients who enrolled in a recommended program did not differ from those who did not enroll based on either levels of agreement or average number of physician–patient agreements for the 7 clinical domains (P's > 0.05). Patients recommended for higher‐intensity programs were perceived by their evaluating physician to have a significantly greater number of problematic clinical domains than those recommended for less intense pain programs.ConclusionThe level of physician–patient agreement regarding the patients’ current difficulties did not appear to influence patients’ decisions to participate in interdisciplinary pain management. Extraneous, nonclinical factors may have had a greater impact on participation in interdisciplinary pain management than physician–patient agreement. Future research should focus on identifying these factors and their impact. Also, studying the impact of physician–patient agreement beyond enrollment status (eg, on successful program completion) may be helpful in potentially enhancing patient outcomes.
Tweeting the Headache Meetings: Cross‐Sectional Analysis of Twitter Activity Surrounding American Headache Society Conferences
Objectives To describe and analyze Twitter activity associated with American Headache Society (AHS) conferences and evaluate the potential for Twitter to promote education and public outreach. Background Many medical and scientific conferences have adopted Twitter as a method of promoting discussion among attendees as well as increasing visibility. Relatively little is known, however, about the composition of conference Twitter activity, the participants, and the impact on broader Twitter discussions. Methods We analyzed Twitter data from 5 AHS conferences held from 2014 to 2016 using their respective hashtags. Using the Symplur Healthcare Hashtags open social media search platform, we gathered data on numbers of tweets, impressions, participants, and mentions during a 10‐day period surrounding each conference, as well as samples of Twitter accounts participating. Prominent accounts were categorized as individual medical professionals, other individuals, host organizations, health‐related organizations, medical centers, and industry by cross‐checking their Twitter profiles and conference registration lists. Larger samples of accounts participating in the 2016 conferences were also obtained and categorized similarly, with individual person accounts classified by conference registration status. A related prominent hashtag (#migraine) was also identified and Twitter usage before, during, and after each conference was analyzed to evaluate the impact of conference activity on broader Twitter conversations. Results Nineteen thousand nine hundred thirty‐six tweets were generated across the 5 conferences, with 11,531 (58%) created by the Top 10 participating accounts in each conference, which were primarily individual medical professionals and host organizations. Thirty‐two million six hundred eighty‐three thousand impressions were generated across the 5 conferences, with 24,656,000 (75%) coming from the Top 10 participants in each, particularly host organizations and other individuals. An average of 331 accounts participated in each conference. The Top 10 mentioned accounts in each conference (consisting of 21 unique accounts with 14 accounts in the Top 10 across multiple conferences, primarily individual medical professionals) received a total of 15,093 mentions. Among 135 unique accounts participating actively in the two 2016 conferences, 39% were individual medical professionals, 38% other individuals, 16% health‐related organizations (including the 2 host organizations), 4% medical centers, and 2% industry. From these samples, 34 of 70 (49%) and 43 of 66 (65%) individual person accounts participating in the Twitter discussion at each conference were not registered conference attendees, indicating substantial outside participation via Twitter. #migraine usage during conferences showed a significant increase from baseline in number of tweets (6080 in a 10‐day period vs 3721, P < .0001) and participants (2332 vs 1830, P < .0001) but the increase was not significant for impressions (30,155 vs 25,361, P = .240). Conclusions Consistent with the dynamics of Twitter conversations on other topics, AHS conference discussions featured a small group of accounts creating the bulk of content, with individual medical professionals and host organizations generating the largest shares of tweets and mentions while host organizations and other individuals produced the most impressions. Participating accounts were mainly individuals and health‐related organizations, with more non‐attendee participants than expected. Conference Twitter activity correlated with a significant increase in #migraine usage, suggesting a perceptible influence on the discussion of health‐related topics beyond the conference itself.
A deep learning‐ and partial least square regression‐based model observer for a low‐contrast lesion detection task in CT
PurposeThis work aims to develop a new framework of image quality assessment using deep learning‐based model observer (DL‐MO) and to validate it in a low‐contrast lesion detection task that involves CT images with patient anatomical background.MethodsThe DL‐MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS‐DA) model and an internal noise component. The CNN was previously trained to achieve the state‐of‐the‐art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL‐MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL‐MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection‐based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two‐alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL‐MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL‐MO and HO.ResultsA statistically significant positive correlation was observed between DL‐MO and HO for the 2AFC low‐contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences.ConclusionsThe proposed DL‐MO is highly correlated with HO in a low‐contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL‐MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.
Down but not out—addressing the scourge of late pulmonary vein reconnection
Objectives/HypothesisDescribe the preoperative evaluation of patients with glottic diastasis who are candidates for endoscopic posterior cricoid reduction (EPCR) and their perioperative and postoperative surgical and voice outcomes, and validate the aerodynamic benefit of EPCR using computation fluid dynamics (CFD)‐based modeling from computed tomography (CT) scans.Study DesignRetrospective case series.MethodsThirteen patients who underwent EPCR were followed from 2013 to 2017. They received a preoperative voice evaluation, microlaryngoscopy and bronchoscopy, dynamic voice CT (performed on patients seen from 2014 to study completion), and postoperative voice evaluation (n = 12). Postoperative inpatient days, complications, and postoperative endoscopic intervention were collected. To validate the aerodynamic benefit of EPCR, CFD modeling was carried out on one patient.ResultsThirteen patients (nine females, nine with intubation injury, and four with post–airway reconstruction dysphonia) underwent EPCR at a mean age of 11.0 years. The mean preoperative and postoperative Pediatric Voice handicap Index scores were 53.8 and 33.8, respectively (P = .006). Mean maximum phonation time preoperatively and postoperatively was 5.3 and 6.7, respectively (P = .04). Of eight patients who underwent preoperative CT imaging, all demonstrated a posterior gap. Modeling demonstrated a change in flow and pressure. The mean hospital stay was 2.4 days. Nine patients underwent postoperative endoscopic intervention, and one experienced a complication that resolved with intervention.ConclusionsPatients who underwent EPCR for dysphonia following intubation or airway reconstruction showed improvements in vocal efficiency, loudness, and perceived voice handicapping. Their hospital stay was brief, with few complications. CFD modeling corroborated these clinical findings. EPCR thus warrants consideration in the management of patients with posterior glottic diastasis.Level of Evidence4 Laryngoscope, 129:S1–S9, 2019
Management of Helicobacter pylori in Patients with Immune Thrombocytopenia
AbstractThere is an association between Helicobacter pylori infection and immune thrombocytopenia (ITP), and few studies have suggested that eradicative treatment of H. pylori infection may improve platelet counts in patients with ITP. Conventional treatments for ITP include immunosuppressive agents, and more recently thrombopoietic agents. However, based on clinical reports of association between H. pylori and ITP, several medical societies increasingly suggest detection and eradication of H. pylori as a treatment for ITP. In this article, we reappraise recent medical literature to determine the effectiveness of platelet response after treatment of H. pylori infection in patients with ITP. We searched two online databases (MEDLINE and Google Scholar) for full articles published between January 2008 and May 2018, and found a total of 11 studies that presented data and outcomes of treatment of H. pylori infection in ITP patients. All the studies administered triple therapy (amoxicillin 500 mg, clarithromycin 250 mg and a proton-pump inhibitor each given twice daily for either 7- or 14-day course) for eradication of H. pylori. Median overall platelet response ranged from 27 to 69.2% with a complete response rate ranging from 0 to 65.4% and a partial response rate ranging from 0 to 29.4%. Although there is variability in the effectiveness between different populations, it appears to be of benefit to ITP patients with concomitant H. pylori infection when treated with triple therapy. However, further studies to understand the pathogenesis of H. pylori-associated ITP is necessary for the development of new therapeutic approaches for ITP.
Go for the Gold: A “Plane” and Simple Technique for Resecting Benign Peripheral Nerve Sheath Tumors
Abstract BACKGROUND Intraneural dissection is a useful technique for achieving gross total resection and preserving functional fascicles for peripheral nerve tumors. Finding the correct tissue plane is the critical step for safe successful enucleation. The authors hypothesized that the yellow color of benign nerve tumors can be used to identify surgical planes. OBJECTIVE To describe a technique to find the correct intraneural dissection plane based on a quantified yellow appearance during resection of benign peripheral nerve sheath tumors. METHODS Intraoperative photographs were reviewed to determine the percentage of yellow that tumors appeared at different phases of surgery. A technique was developed to quantitatively measure the amount of yellow using Photoshop. Previously published journal articles containing color images of peripheral nerve tumor resections were also analyzed for yellow tumor color. RESULTS There were 24 patients with suitable images to permit measurement of color for 3 steps of the procedure. The average percentages of yellow for tumor exposure, tumor resection, and removed specimen were 36.5%, 59.1%, and 80.4%, respectively ( P < .001). Three publications were found that contained high-quality images of at least 2 phases of the surgery with average yellow content of 47.4% and 84% ( P < .01). CONCLUSION The simple observation that a benign nerve sheath tumor is yellow in color can be used to guide surgical resection and achieve excellent outcomes. Intraneural dissection through the pseudocapsule should be pursued layer by layer until a yellow true capsule is found. By sparing functional nervous tissue within the pseudocapsule, this technique may lead to improved neurological outcomes.
Plerixafor alone for the mobilization and transplantation of HLA-matched sibling donor hematopoietic stem cells
Abstract Plerixafor, a direct antagonist of CXCR4/stromal-derived factor 1, can safely and rapidly mobilize allografts without the use of granulocyte colony-stimulating factor (G-CSF). We conducted a phase 2, multicenter, prospective study of plerixafor-mobilized HLA-identical sibling allografts for allogeneic hematopoietic cell transplantation in recipients with hematological malignancies. Donors (n = 64) were treated with subcutaneous plerixafor (240 µg/kg) and started leukapheresis (LP) 4 hours later. The primary objective was to determine the proportion of donors who were successfully mobilized: defined as collection of ≥2.0 × 106 CD34+ cells per kilogram recipient weight in ≤2 LP sessions. Recipients subsequently received reduced intensity (RIC; n = 33) or myeloablative (MAC; n = 30) conditioning. Sixty-three of 64 (98%) donors achieved the primary objective. The median CD34+ cell dose per kilogram recipient weight collected within 2 days was 4.7 (0.9-9.6). Plerixafor was well tolerated with only grade 1 or 2 drug-related adverse events noted. Bone pain was not observed. Plerixafor-mobilized grafts engrafted promptly. One-year progression-free and overall survivals were 53% (95% confidence interval [CI], 36% to 71%) and 63% (95% CI, 46% to 79%) for MAC and 64% (95% CI, 47% to 79%) and 70% (95% CI, 53% to 84%) for RIC recipients, respectively. Donor toxicity was reduced relative to G-CSF mobilized related donors. This is the first multicenter trial to demonstrate that, as an alternative to G-CSF, plerixafor rapidly and safely mobilizes sufficient numbers of CD34+ cells from matched sibling donors for HCT. Engraftment was prompt, and outcomes in recipients were encouraging. This trial was registered at clinicaltrials.gov as #NCT01696461.
Response: Goniotomy Using the Kahook Dual Blade in Severe and Refractory Glaucoma: 6-Month Outcomes