Medicine Correspondence Blog

The Medicine Correspondence Blog allows authors to post Letters to the Editors, Reviews, and other editorial writings that are not considered original research.

Monday, June 25, 2018

We commend Peng et al.[1] for devising "stage operation" for placenta accreta. They placed a pressure band, reduced the blood flow, and removed the placenta cephalad to it, with the remaining placenta left in situ. Although 2 of 19 patients required hysterectomy after cesarean section (CS), the uterus was preserved in remaining 17. Four patients required dilatation and evacuation after CS and the placenta was resolved or discharged in remaining 13.


International Federation of Gynecology and Obstetrics recently recommended the usage of the terminology of "placenta accreta spectrum (PAS) disorders (creta, increta, percreta)",[2,3] thereby avoiding the ambiguity of the term "accreta", which has been used in a double sense, narrow and wide. Sub-classification of PAS (creta, increta, percreta) should be made only by histological examination after hysterectomy. When hysterectomy was not performed, "PAS" (without sub-classification) should be used. Thus, Peng et al. performed this two-stage surgery to patients with PAS. We have clarification and concern.


Our clarification is: PAS surgery can be categorized into four:[4,5] i) forcible placental removal (extirpative approach), ii) partial resection of the uterine wall, with the PAS-placenta adhering to the uterine wall to be resected, iii) placenta left in situ approach, iv) hysterectomy either cesarean or intentionally delayed. The uterus is preserved in i)-iii) and resected in iv). Peng et al.'s procedure is a combination of i) and iii). The removal of the partial placenta reduces the uterine volume, accelerating uterine contraction and thus involution. Reduction of the total placental volume left in situ may reduce the possibility of placental infection. Nobody, to our knowledge, has proposed this concept. This procedure may become an option for PAS surgery.


Our concern is: this partial placental removal is performed effectively and successfully when i) the placental part cephalad to the pressure band is relatively thin, ii) this part is removable, and iii) its removal actually reduces the volume of the remaining placenta (caudal to the band). Figure 1a illustrates this scenario 1. In the scenario 2 (Figure 1b), contrary to the scenario 1, the placenta exists just adjacent to the internal ostium in a lump and even the "cephalad" part also has severer PAS (diffuse increta/percreta). Then, one can hardly separate the "cephalad" part of the placenta, and even if one can do it, the removed part of the placenta is small. This "forcible" placental removal may cause marked bleeding even with the use of the pressure band or some hemostatic techniques. In short, if the removal part is large to the extent that it reduces the uterine/placental volume, partial removal may be effective, whereas if its removal may not reduce their volume and may cause marked bleeding, partial removal is not effective but rather dangerous. Although scenario 1 vs. 2 illustrates an extreme example, less experienced obstetricians may not be able to discern whether the placenta of an individual patient belongs to type (scenario) 1 or 2. This should be clarified for Peng et al.'s procedure to be reproducible. It regards the indication of this new surgery. 



Shigeki Matsubara, Hironori Takahashi

Director and Professor, Department of Obstetrics and Gynecology,



[1] Peng Q, Zhang W, Liu Y. Clinical application of stage operation in patients with placenta accreta after previous caesarean section. Medicine (Baltimore). 2018;97:e10842.

[2] Jauniaux E, Silver RM, Matsubara S. The new world of placenta accreta spectrum disorders. Int J Gynaecol Obstet. 2018;140:259-60.

[3] Matsubara S, Jauniaux E. Placenta accreta spectrum disorders: A new standardized terminology better defining the condition. J Obstet Gynaecol Res. 2018 Apr 19. doi: 10.1111/jog.13649.

[4] Sentilhes L, Goffinet F, Kayem G. Management of placenta accreta. Acta Obstet Gynecol Scand. 2013;92:1125-34.

[5] Matsubara S, Takahashi H. Intentional placental removal on suspicious placenta accreta spectrum: still prohibited? Arch Gynecol Obstet. 2018;297:1-2.


Figure legend

Figure 1. Schema of two extreme types of placenta accreta spectrum (PAS) disorders.

  1. The placenta widely covers the anterior uterine wall. The placenta cephalad to the pressure band is without PAS or less severe PAS (creta). The removal of this placental part is easy and may reduce the uterine and remaining placental volume.

    2. The placenta locates adjacent to the internal ostium in a lump with diffuse increta or percreta. The removable placental parts are        small and their forcible removal may cause marked bleeding.   

Monday, June 4, 2018

Venous thromboembolism including cerebral venous thrombosis is a known complication of cancer. Pancreatic cancer patients have traditional risk factors for stroke including but not limited to hypertension, diabetes, smoking, and atrial fibrillation. Chan et conducted a nation-based retrospective cohort study in Taiwan including 7479 pancreatic cancer patients and propensity score matched 29,916 controls. The cumulative incidence of all strokes was higher in pancreatic cancer group (5.5% vs 3.7%, p<0.001). The highest incidence appeared to be within first six months of diagnosis.

We evaluated what proportion of stroke in the entire cohort is attributable to pancreatic cancer. We calculated population-attributable risk percent (PAR%). PAR% for all strokes is equal to incidence of stroke in total cohort subtracted to incidence of stroke among controls divided by incidence of stroke in total cohort. In this case, PAR%=8.21% suggests that 8.21% of all strokes in whole cohort is attributable to pancreatic cancer. PAR% for ischemic and hemorrhagic strokes in whole cohort is equal to 8.93% and 6.43%, respectively. Another nationwide study conducted in Sweden similarly reports increased risk of ischemic stroke in pancreatic cancer patients particularly in metastatic settings. The article by Navy et al provides further support to the above data. The authors reported increased three-month cumulative incidence of stroke in patients with pancreatic cancer compared to matched controls in SEER database review (3.4% vs 1.3%, respectively, P<0.0001).

Possible mechanisms of ischemic stroke in the absence of conventional risk factors include hemostatic system imbalance caused by inherited or acquired coagulation factors or blood vessel specifically arterial endothelial dysfunction. Increased tumor burden results in tumor emboli and production of inflammatory mediators similar to IL 6 and CRP which lead to endothelial cell damage and release of tissue factor initiating thrombosis. Location of cancer in pancreas considered a very high risk for prediction of venous thromboembolism based on Khorana score.

In conclusion, the findings are important to consider a pancreatic cancer as a non-conventional risk factor for the stroke. This warrant a close follow up of pancreatic cancer patients, evaluating and strictly controlling the traditional risk factors, and identifying those patients who are at very high risk for stroke. In those pancreatic cancer patients who are on anticoagulation due to venous thromboembolism prevention we need to carefully outweigh benefit versus risk due to increased risk of hemorrhagic stroke. ​


 Chan P-C, Chang W-L, Hsu M-H, et al. Higher stroke incidence in the patients with pancreatic cancer: A nation-based cohort study in Taiwan. 97(11).

 Navi BB, Reiner AS, Kamel H, Iadecola C, Elkind MS, Panageas KS, DeAngelis LM. Association between incident cancer and subsequent stroke. Ann Neurol. 2015;77:291–300. [PMC free article][PubMed]

 Zoller B, Ji J, Sundquist J, et al. Risk of haemorrhagic and ischaemic stroke in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012;48:1875–83.

 Correspondence to :


Lilit Karapetyan, MD, MS candidate 2019​


Monday, June 4, 2018

We read the article entitled 'Modified transprepancreatic septotomy reduces postoperative complications after intractable biliary access" by Zhong et al with great interest (1). The authors found that the bile duct cannulation success rate was the same for transpancreatic septotomy (referred to as the septotomy group) with guide-wire and  transprepancreatic septotomy with pancreatic duct stent (modified septotomy group) but the former has a lower complication rate, especially concerning pancreatitis.

It is well known that difficult cannulation itself is associated with an increased risk of post ERCP complications. In cases of biliary cannulation failure, several supplementary techniques of precutting have been recommended to facilitate access to the bile duct. Transpancreatic septotomy is a relatively new precut technique and has been shown to have a high bile duct cannulation success rate (90-95.9 %) and a low complication rate (2.9-14.3 %) (2-4).

Although the authors reported no increased risk of perforation in these studies, we think that the attempts for bile duct cannulation with a guide wire after transpancreatic septotomy and pancreatic stent placement can increase the risk of perforation or dissection. Recently, we experienced two patients with perforation or dissection. In the first patient, we performed transpancreatic septotomy over a guide-wire and inserted a 5-Fr pancreatic stent into the pancreatic duct. Later on we realised retroperitoneal free air after a few attempts for bile duct cannulation with a guide wire loaded sphincterotome (Type-II perforation) (Figure-1). In the second patient, the guide-wire was passed to the paracholedochal soft tissue during the cannulation attempts after transpancreatic septotomy and air was noticed around the common bile duct on the scopic images (paracholedochal dissection) (Figure-2). We placed fully-covered self expandable metal stents into the common bile duct in both of the patients and they were followed conservatively and discharged without any problem. We hypothesized that the septum seperating the common bile duct and pancreatic duct was ablated by transpancreatic septotomy, and as a result of loss of tissue integrity, the guide wire was passed through the soft tissue or the common bile duct, causing perforation or dissection.

Taking this into consideration, we would like to warn the endoscopists about the potential risk of perforation or paracholedochal dissection after transpancreatic septotomy.

Figure 1:


Figure 2:



Ahmet Tarik Eminler1, Aydin Seref Koksal1, Erkan Parlak2

1Sakarya University Faculty of Medicine, Department of Gastroenterology, Sakarya, Turkey

2Hacettepe University Faculty of Medicine, Department of Gastroenterology, Ankara, Turkey


The authors have no conflict of interest


Correspondence Author:

Ahmet Tarik Eminler,MD

Sakarya University Faculty of Medicine, Department of Gastroenterology, Korucuk Kampus, Sakarya, Turkey




  1. Zhong H, Wang X, Yang L, et al. Modified transprepancreatic septotomy reduces postoperative complications after intractable biliary access. Medicine (Baltimore). 2018 Jan;97(1):e9522
  2. Wang P, Zhang W, Liu F, et al. Success and complication rates of two precut techniques, transpancreatic sphincterotomy and needle-knife sphincterotomy for bile duct cannulation. J Gastrointest Surg. 2010 Apr;14(4):697-704
  3. Zang J, Zhang C, Gao J. Guidewire-assisted transpancreatic sphincterotomy for difficult biliary cannulation: a prospective randomized controlled trial. Surg Laparosc Endosc Percutan Tech. 2014 Oct;24(5):429-33.
  4. Huang C, Kung J, Liu Y, et al. Use of double wire-guided technique and transpancreatic papillary septotomy in difficult ERCP: 4-year experience. Endosc Int Open. 2016 Oct;4(10):E1107-E1110

Friday, May 18, 2018

We read with great interest the study from Werner J. and Coll on hydroxyethyl  starch (HES) 6% or 10% or balanced crystalloid use for goal-directed hemodynamic algorithm in pancreatic surgery.[1] As reported in the text, this is a sponsored-initiated study funded by B. Braun Melsungen AG (that is one of the producers of HES containing solutions). The Authors report that the study was "interrupted for futility" when a total of 63 patients have been enrolled (20 in HES 10%, 22 in HES 6% and 21 in balanced crystalloid treatment), while the initial sample size calculation lead to an estimate of 76 patients per group, according the suggestion from a Independent Data Monitoring Committee (IDMC).

We wonder how large was the role that the sponsor had in the IDMC. This is especially relevant since the reported results demonstrate that HES 10% use is associated with an increased acute kidney injury (AKI) and are "a step away" (in terms of statistical significance) from providing evidence on HES 6%-associated AKI. 

As additional point, the listed safety and efficacy tested end points (i.e. total amount of infused fluids and the time until fully on oral solid diet) were not associated with any beneficial HES effect and are disproportionally far from the ongoing concerns related with HES use in the perioperative setting that include AKI and excess of bleeding.[2,3] Since the HES use seems to be challenged by a growing number of possible associated drawbacks, isn't it time to produce clinical trials where "heavy end points" (mortality, functional follow up status at long term, etc) are tested in order to detect if there is a clear clinical benefit in HES use?

Author Correspondence:

Giordano Giovanni, MD, Bilotta Federico, PhD,​



[1]        Werner J, Hunsicker O, Schneider A, et al. Balanced 10% hydroxyethyl starch compared with balanced 6% hydroxyethyl starch and balanced crystalloid using a goal-directed hemodynamic algorithm in pancreatic surgery: a randomized clinical trial. Medicine (Baltimore) 2018 Apr; 97(17): e0579.

[2]        Hartog CS, Natanson C, Sun J, Klein HG, Reinhart K. Concerns over use of hydroxyethyl starch solutions. Brit Med J 2014; 349: g5981.

[3]        Wilson J. Erroneous conclusions about the safety of hydroxyethyl starch 130/0.4 in paediatric cardiac patients? Anaesthesia 2018 May; 73(5): 649-650.

Friday, May 18, 2018

We have recently read the original article entitled "Point-of-care ultrasound versus auscultation in determining the position of double-lumen tube" by Huet al. with great interest.[1]In their study, Hu et al. compared  auscultation and ultrasound (US) methods to confirm the position of double-lumen tubes (DLTs), and reported that US is superior to auscultation method. However, we noticed that Hu et al. defined the motion mode (M-mode) US image of a non-ventilated healthy lung as "barcode (stratosphere) sign" after the endobronchial intubation and showed as a figure.[1] Even so, we consider that the image presented by Hu et al. should have been defined as "lung pulse (LP)", but not as "barcode sign".[2-5]

In recent studies, US has been demonstrated to be used in the confirmation of endotracheal and endobranchial intubation.[3-6] Having separated the right and left lungs after the endobranchial intubation, a few ultrasonographic signs can be used in the determination of the ventilated and non-ventilated lungs.[3,4,7] The most important key sign is the "lung sliding (LS)" in the brightness mode (B-mode) US image. LS is a sign of healthy interpleural cavity determining the movement of visceral pleura to parietal pleura during respiration. If this investigation is carried out on M-mode in a normal ventilated lung, an image called "seashore sign" is formed.[2]

While LS disappears in the non-ventilated healthy lung, the heart beats are transmitted to the immobile lung and the image in the M-mode is called as LP. The vibrations reflected vertically to the pleura form an image in the view of block synchronized with the heart beats. LP sign has the rates of 93% sensitivity and 100% specifity in the selective left side intubation.[2] Although the same image is observed in the consolidated motionless lung or as a result of breath holding, LP is seen more frequently in the selective left side intubation depending on its proximity to the heart. However, if the lung is well-ventilated, LS invalidates LP.

Pneumothorax is characterized by the absence of LS and LP marks.[8] The dynamic motion, a sign of ventilation, disappears with the separation of two pleural layers. The heart beats cannot be transmitted to the pleura due to air layer.The air layer between visceral and parietal pleuras prevents the reflection of both horizontal and vertical movements. Parallel-line image composed of the sonographic view of the skin and muscle layers of thoracic wall on parietal pleura. The sand-like appearance under the pleural line is replaced by this parallel lines with pneumothorax. This image obtained characteristically to the pneumothorax on M-mode is termed as "barcode (stratosphere) sign" [9,10]  .

In various studies with similar findings where US was used in the confirmation of the place of DLTs, the image formed when there was no LS was called as LP, but not as barcode (stratosphere).[2-5] As a consequence, while LP image is obtained when LS disappears in a non-ventilated normal lung due to any reason, the barcode (stratosphere) sign takes place when pneumothorax develops (Figure 1).


Figure 1: A: Seashore sign, B: The lung pulse, C: Barcode (stratosphere) sign.

 lte determination.jpg



Corresponding Author:

Funda Gok MD     


Alper Kilicaslan MD, DESA    ​



  1. Hu WC, Xu L, Zhang Q, et al. Point-of-care ultrasound versus auscultation in determining the position of double-lumen tube. Medicine.  2018 ; 97 (13).
  2. Lichtenstein DA, Lascols N, Prin S, et al. The lung pulse an early ultrasound sign of complete atelectasis. Intensive Care Med. 2003; 29: 2187-92.
  3. Álvarez-Díaz N, Amador-García I, Fuentes-Hernández M, et al. Comparison between transthoracic lung ultrasound and a clinical method in confirming the position of double-lumen tube in thoracic anesthesia. A pilot study. Rev Esp Anestesiol Reanim. 2015; 62(6):305-312.
  4. Sustić A1Protić ACicvarić Tet al. The addition of a brief ultrasound examination to clinical assessment increases the ability to confirm placement of double-lumen endotracheal tubes. J Clin Anesth 2010; 22(4): 246-9.
  5. Nam JS, Park I, Seo H, et al. The use of lung ultrasonography to confirm lung isolation in an infant who underwent emergent video-assisted thoracoscopic surgery -a case report. Korean J Anesthesiol. 2015; 68(4):411-4.
  6. Sim SS, Lien WC, Chou HC, et al. Ultrasonographic lung sliding sign in confirming proper endotracheal intubation during emergency intubation. Resusciation 2012; 83: 307-12.
  7. Blaivas M, Tsung JW. Point-of-care sonographic detection of left endobronchial main stem intubation and obstruction versus endotracheal intubation. J Ultrasound Med 2008; 27: 785-9.
  8. Volpicelli G. Sonographic diagnosis of pneumothorax. Intensive Care Med. 2011; 37: 224-232.
  9. Chuang TJ, Lai CC. Sonographic barcode sign of pneumothorax. QJM. 2017; 110 (8): 525-526.
  10. Lichtenstein DA. Lung ultrasound in acute respiratory failure an introduction to the BLUE-Protocol. Minerva Anestesiol. 2009; 75: 313-7.