OPINION
Opinion By
Appellant Ethicon Endo-Surgery, Inc. appeals the trial court’s judgment rendered in favor of appellee Celia Gillies, individually, and as next friend of Edward Allan Ortega, a minor and heir to the Estate of Rebecca T. Castaneda. We reverse and render judgment in favor of appellant.
Background
Rebecca Castaneda had a body mass index of over 50, which is medically referred to as “super-obesity,” and suffered from diabetes and high blood pressure related to her weight. In May 2005, Ms. Castaneda opted to undergo gastric bypass surgery to combat her obesity. Dr. John Mason began Ms. Castaneda’s surgery as a laparoscopic procedure and selected a *208LONG45A Endocutter1 made by appellant with the blue cartridge to seal off and divide the stomach. The record reflects that the LONG45A is made to work with different staple cartridges, which are col- or-coded, depending on the closed-staple height: white 1.0 mm; blue 1.5 mm; and green 2.0 mm. Each LONG45A comes in a box with Instructions for Use (“IFU”) that: (1) give suggestions as to which staple size to use; and (2) provide contraindications for use of the device. The IFUs warn surgeons not to use LONG45A with the green cartridge on any tissue that requires excessive force to compress to 2.0 mm or on tissue that compresses easily to less than 2.0 mm. The IFUs provide similar warnings for the blue cartridges, which are for tissue that compresses to 1.5 mm.
On Dr. Mason’s first firing with the blue cartridge, the staples did not completely form or close, creating a hole in Ms. Castaneda’s stomach. Dr. Mason then decided to convert from laparoscopic surgery to an open surgery, whereby a LONG45A with a green cartridge was used to close the gastrotomy on both the pouch and remnant side. Dr. Mason observed the green staples created solid staple lines.
Following surgery, Ms. Castaneda progressed normally for two days and appeared to be doing well. Three days following surgery, however, she developed increased abdominal pain following some retching and vomiting. She further developed an infection and rapid heart rate, had diminished urine output, and showed signs of sepsis with no clear source. In view of these symptoms, Dr. Mason decided to do exploratory surgery on the morning of the fourth day.
During that surgery, Dr. Mason found a pinhole leak in the green staple line on the remnant side. Bile had escaped through this, forming a biloma and causing infection and fever. Dr. Mason did not know what caused the pinhole leak. He washed out Ms. Castaneda and sutured the remnant side to repair the leak. Following the exploratory surgery, Ms. Castaneda seemed to be doing well and was transferred from the ICU to a regular room.
Seven days after the exploratory surgery, Ms. Castaneda became unresponsive and went into cardiac arrest. Dr. Mason did not know what caused her to become unresponsive and took her into the operating room for a third surgery because her abdomen was distended. He found nothing in her abdomen indicating a cause, only some blood from an apparent trauma. Ms. Castaneda died that same day. An autopsy was performed by Dr. Chen, where it was determined that the immediate cause of Ms. Castaneda’s death was a pulmonary thromboembolism, which is a blockage in the blood vessels of the lungs caused by a blood clot. Dr. Chen observed the green staples were not malformed.
Appellee filed suit against appellant. The first trial resulted in a mistrial after a hung jury. In the second trial, the case proceeded under appellee’s second amended petition, in which she alleged both (1) strict liability for appellant’s defective design, manufacture, assembly and marketing of the surgical stapler and/or staples and (2) negligence in appellant’s design, manufacture, assembly and marketing of the surgical stapler and/or staples, among other causes of action.2
*209During the course of the second trial, appellee represented to the trial court she was not asserting a manufacturing defect.3 Furthermore, the record before us reflects appellee orally non-suited her “design defect claim” during a jury recess.
At the end of trial, the court charged the jury and submitted nine questions. In the first question, the trial court asked the jury to determine the strict liability question as to whether there was “a defect in the marketing of the product at the time it left the possession of [appellant] that was a producing cause of the occurrence in question.” The jury answered in the negative (in favor of appellant) to both the Long45A Stapler with the blue cartridges and the green cartridges. The second question asked the jury whether “the negligence, if any, of [appellant] proximately caused the occurrence in question.” The jury answered in the affirmative (in favor of ap-pellee) and awarded damages to appellee, totaling $320,000. The jury answered the remaining questions in favor of appellant, i.e. there was no misrepresentation by appellant and no gross negligence on the part of appellant and, therefore, no damages awarded for those claims. This appeal ensued.
Analysis
Appellant raises five issues on appeal. First, appellant alleges there is no evidence to support the jury’s negligence finding, specifically with regard to standard of care and proximate cause. Second, appellant claims appellee’s failure to prove a marketing defect precludes her negligence claim, which was based solely on a complaint about marketing. Third, it argues the trial court abused its discretion in excluding evidence of the conduct of Ms. Castaneda’s surgeon. Fourth, appellant contends the trial court abused its discretion in its refusal to instruct the jury on sole proximate cause and new-and-independent cause, because such instructions were supported by the evidence. Finally, it argues the trial court erred in not limiting appellee’s recovery of medical expenses to those actually paid or incurred.
We turn to appellant’s first issue, in which it contends there is no evidence to support the jury’s negligence finding on question 2. In their briefs before us, the parties disagree which negligence claims were actually submitted to the jury under question 2. Appellee contends that, in addition to her marketing claim, her negligent design claim went to the jury. Appellee thus argues because negligent design went to the jury and there was evidence of negligent design, the jury properly found appellant was negligent in question 2. We disagree.
As noted above, appellee informed the trial court that she had dropped her manufacturing, and thus, assembly defect claims. Appellee also non-suited her design defect claim during a jury recess when her counsel stated, “[Plaintiffs going to nonsuit its design defect claim.” The right to a non-suit has a long and well-established tradition in Texas jurisprudence. Rule 162 provides that a non-suit may be taken until all the evidence has been presented. See Tex.R. Civ. P. 162. To take a non-suit the party *210can file either a written motion to dismiss or make an oral announcement of the non-suit in open court. Greenberg v. Brookshire, 640 S.W.2d 870, 872 (Tex.1982). The movant may choose to take a non-suit as to some claims and parties -without non-suiting the rest of the claims or parties. See Tex.R. Civ. P. 162, 168. See also Cook v. Nacogdoches Anesthesia Grp. LLP, 167 S.W.3d 476, 482 (Tex.App.-Tyler 2005, no pet.). The party requesting a non-suit has an absolute right to a non-suit at the moment the motion is filed. Greenberg, 640 S.W.2d at 872; McGowen v. Huang, 120 S.W.3d 452, 461 (Tex.App.-Texarkana 2003, pet. denied). The granting of a non-suit is merely ministerial and gives the trial court no discretion. Greenberg, 640 S.W.2d at 872; McGowen, 120 S.W.3d at 462.
Appellee contends that the non-suit of her strict-liability “design defect” claim did not include a non-suit of her “negligent design” claim. Although, in isolation, we consider appellee’s non-suit of her “design defect claim” vague as to whether it included both her strict-liability design defect and negligent design claims, it is clear from the record that the trial court and the parties understood the non-suit to include both the strict-liability design defect and negligent design claims.
Following appellee’s non-suit, her counsel attempted to exclude evidence of a video clip, arguing, “One thing I want to point [out] is because we have dismissed our design defect claims, there is no more issue about the design defect with respect to tissue thickness, the ability to measure force, the fact that it clamps down too much tissue, the larger stapler.” The trial court agreed and asked defense counsel “why show this particular video if that’s all gone[?]” Appellee’s trial counsel reiterated the objection, stating the video clip was “no longer relevant because we’ve dropped the design defect claim and it’s unfairly prejudicial because it goes to design defect and it goes to the surgeon and doctor negligence.... And this is not a marketing video.”
In addition, appellant’s trial counsel later requested the trial court “get the jury advised that the issue of design defect is no longer in this case because I think they are now wondering why I brought my engineer ...” The trial court noted “this is a simple lawsuit at this point” and asked appellant’s counsel to only “ask marketing questions.” Further, the trial court told counsel that asking the engineer “about the surgical judgment and how they choose cartridges, first of all, is ... beyond what this case is not about. It’s about marketing. Let’s please, limit it to that.”
Before reading the charge, the court told the jury: “The trial was in fact shortened at our break yesterday, and you are no longer needed to decide the design defect issue.” Finally, appellant’s trial counsel argued in closing argument: “It’s interesting to note that there is not a question in the charge with regard to whether or not our product is designed defectively. That issue is no longer in this case....”
Appellee never objected to the statements made during the course of trial, which limited the case to a marketing case, and thus we cannot review that complaint now. Tex.R.App. P. 33.1. Further, because a party is restricted on appeal to the theory on which the case was tried, we will not affirm the trial court’s judgment based on a legal theory not presented to the trial court and to which the other party had no opportunity to respond. Victoria Gardens of Frisco v. Walrath, 257 S.W.3d 284, 290 (Tex.App.-Dallas 2008, pet. denied); Vaughn Bldg. Corp. v. Austin Co., 620 S.W.2d 678, 683 (Tex.Civ.App.-Dallas *2111981), aff'd as modified, 643 S.W.2d 113 (Tex.1982). Based upon the statements of both counsel and the trial court, including its limitation of evidence due to its stated understanding that the case remained only about marketing, we conclude appellee non-suited both her strict liability design defect claim and her negligent design claim. Therefore, the only remaining negligence claim asserted against appellant was a claim for negligent marketing.4
Having determined the only claim before the jury was the negligent marketing claim, we now turn to the remainder of appellant’s first issue, asking us to determine whether appellee failed to present evidence in support of her claim for negligent marketing, specifically with regard to standard of care and proximate cause. Whether a plaintiff seeks recovery because of negligence or strict liability in tort, the burden is on the plaintiff (appel-lee in this case) to prove that the injury resulted from a defect in the marketing of the product. See Ford Motor Co. v. Miles, 141 S.W.3d 309, 317 (Tex.App.-Dallas 2004, pet. denied). To prevail on her negligent marketing claim, as opposed to a strict liability marketing defect claim, the appel-lee was required to establish four elements: 1) a duty by appellant to act according to an applicable standard of care; 2) a breach of the applicable standard of care; 3) an injury; and 4) a causal connection between the breach of care and the injury. Cobb v. Dallas Fort Worth Med. Ctr.-Grand Prairie, 48 S.W.3d 820, 824-825 (Tex.App.-Waco 2001, no pet.) (citing Denton Reg. Med. Ctr. v. LaCroix, 947 S.W.2d 941, 950 (Tex.App.-Fort Worth 1997, pet. dism’d by agr.)). Although our review of Texas case law demonstrates a consensus that expert testimony is required in the context of strict liability marketing defect claims, no Texas court has addressed whether expert testimony is required in negligent marketing cases like the one we have before us. See, e.g., Nissan Motor Co., Ltd. v. Armstrong, 145 S.W.3d 131, 137 (Tex.2004); DeGrate v. Exec. Imprints, Inc., 261 S.W.3d 402, 410-11 (Tex.App.-Tyler 2008, no pet.); Georgia-Pacific Corp. v. Stephens, 239 S.W.3d 304, 321 (Tex.App.-Houston [1st Dist.] 2007, pet. denied); Ethicon Endo-Surgery, Inc. v. Meyer, 249 S.W.3d 513, 516-17 (Tex.App.-Fort Worth 2007, no pet.) (all strict liability marketing defect cases).
However, the Texas Supreme Court has determined that expert testimony is necessary to establish the standard of care “when the alleged negligence is of such a nature as not to be within the experience of the layman.” See FFE Transp. Servs., Inc. v. Fulgham, 154 S.W.3d 84, 90-91 (Tex.2004). When the conduct at issue involves the use of specialized equipment and techniques, expert testimony must establish both the standard of care and a •violation of that standard. See, e.g., Simmons v. Briggs Equip. Trust, 221 S.W.3d 109, 114 (Tex.App.-Houston [1st Dist.] 2006, no pet.) (maintenance and service of TrackMobile vehicle); Hager v. Romines, 913 S.W.2d 733, 734-35 (Tex.App.*212-Fort Worth 1995, no writ) (aerial application of herbicide to crops); Parkway Co. v. Woodruff, 857 S.W.2d 903, 919 (Tex.App.Houston [1st Dist.] 1993) (engineering design of retaining wall), aff'd as modified, 901 S.W.2d 434 (Tex.1995). Typically, Texas law also requires expert testimony to establish the standard of care in medical cases. See Roark v. Allen, 633 S.W.2d 804, 809 (Tex.1982).
Because we conclude the standard of care in marketing a specialized medical device requiring specialized technique for use is not within the experience of laymen, we must also conclude expert testimony was required to prove negligent marketing of such a device as in this case. See Fulgham, 154 S.W.3d at 90-91. Appellee was thus required to offer expert testimony that appellant failed to exercise ordinary care in the marketing of the LONG45A. See Lozano v. H.D. Indus., Inc., 953 S.W.2d 304, 314 (Tex.App.-El Paso 1997, no writ). In her brief, appellee directs us to the testimony of her expert witness, Dr. William Hyman, in support of her negligent marketing claim. We, therefore, look to Dr. Hyman’s testimony to determine whether there was any evidence of the appropriate standard of care for the marketing of the LONG45A and whether Ethicon’s conduct met that standard. See Fulgham, 154 S.W.3d at 91; Simmons, 221 S.W.3d at 114-15 (expert testimony required where industry standards were not within a layperson’s knowledge); Parkway, 857 S.W.2d at 919 (plaintiff failed to offer expert testimony that the engineering design did not meet professional standards).
Over the course of his testimony, Dr. Hyman, a biomedical engineer, testified the LONG45A was defectively designed because it does not measure tissue thickness, does not measure force, locks down on tissue that is too thick, needs a larger staple option, and has a misleading click. In Dr. Hyman’s opinion, the LONG45A is “defectively designed with respect to bar-iatric surgery.” However, when asked whether the LONG45A was negligently marketed, Dr. Hyman responded, “[t]he overriding issue here is that as designed it’s inappropriately designed for bariatric surgery because of the too small staples and the inability to follow the instructions.”
Dr. Hyman testified the marketing of a product “includes training the users so that they can use your product safely as you hopefully intended it to be used.” According to Dr. Hyman, the proper use of warnings is for “what’s left over when you’re done, not to use warnings to try to compensate for bad design.” Appellee’s attorney asked Dr. Hyman what “types of information” a company such as appellant “should provide to the end user of a product” such as the LONG45A. Dr. Hyman responded “an appropriate device with appropriate instructions.” The only specific complaint Dr. Hyman had about the instructions given by appellant regarding the LONG45A was that a physician should not use “excessive force.” Dr. Hyman testified the instructions failed to define “excessive force” and the LONG45A could not measure the force being applied. Dr. Hy-man testified a physician could tell excessive force was being used only by “[h]ow it feels in his hand” and whether “he’s having to squeeze harder than he’s use to.” This, again, appears to be a complaint about the design of the LONG45A, rather than the marketing of the device.
In Dr. Hyman’s opinion, the LONG45A was “actively promoted for a use for which it was inappropriate and is inappropriate and they knew that.” Dr. Hyman pointed to appellant’s internal documents that demonstrated appellant was aware of a number of complaints, beginning as early *213as 1999, about malformed staples from the EST45, a predecessor to the LONG45A. Appellant also sent a safety alert to physicians in 2000 about the problems with the EST45. Dr. Hyman opined that this was “trying to fix design with a letter.” Dr. Hyman testified appellant was aware the EST45, and its successor the LONG45A, could not be used on tissue more than two millimeters thick, but failed to provide larger staples or prevent the device from operating if the tissue was too thick. In Dr. Hyman’s opinion, appellant should have either stopped marketing the EST45, and by implication the LONG45A, and “contraindicated it for bariatric surgery” or “warned people not to use it under those conditions, and if they wanted to be in that market, you know, design the device that was appropriate.”
Dr. Hyman clearly believed the LONG45A, as designed, should not be used in bariatric surgery. His opinions, however, were premised on the LONG45A being inappropriate for use on gastric tissue more than two millimeters thick. He admitted that a study performed by appellant in 2004 indicated ninety percent of the population had gastric tissue less than two millimeters thick and did not offer an opinion that the LONG45A was inappropriate for use on gastric tissue less than two millimeters thick. He also admitted appellant warned physicians not to use the LONG45A on tissue that was more than two millimeters thick.
In Dr. Hyman’s opinion, because the LONG45A was not appropriate for use on gastric tissue more than two millimeters thick, it should not have been marketed for use in bariatric surgery. Simply stating that a product was defectively designed for use in certain situations and, therefore, should not have been marketed at all, does not establish a standard of ordinary care applicable to the marketing of the product for use in other situations. Dr. Hyman did not testify the standard of care applicable to the industry required appellant to give any warning relating to the use of the LONG45A on tissue less than two millimeters thick that was not given. Further, Dr. Hyman admitted appellant warned physicians not to use the LONG45A on tissue more than two millimeters thick. We conclude appellee failed to provide expert testimony establishing either the appropriate standard of care for her negligent marketing claim or that appellant failed to comply with that standard. Therefore, we affirm appellant’s first issue.
Based upon our determination of appellant’s first issue, we need not address its remaining issues. Tex.R.App. P. 47.1. We reverse the judgment of the trial court and render judgment in favor of appellant.